Defined benefit pension schemes The impact on FTSE350 company accounts at 31 December 2011 June 2012
Defined benefit pension schemes The impact on FTSE350 company accounts at 31 December 2011 Contents Headline results 4 The story of 2011 5 Discount rate 7 Inflation 10 Life expectancy 13 Expected return on assets 15 Future changes to IAS19 17 Appendix 20 Introduction Welcome to the first Towers Watson report looking into the disclosure of defined benefit pension schemes in company accounts. This report comes at a time of difficult economic conditions for many pension schemes and their corporate sponsors. The report focuses on the disclosures of FTSE350 companies and provides an in-depth analysis of the factors which influence the size of pension scheme deficits reported at 31 December 2011. 192 FTSE350 companies have financial years ending 31 December 2011. 170 of these had published their accounts at the time that this report was prepared, of which 110 have a defined benefit pension scheme. Our analysis focuses on these companies. Towards the end of the report we also consider forthcoming changes to the accounting standard IAS19 and the impact these may have on a company s disclosures. We hope that you find this report informative. If you would like more information, please contact: Neil Crombie +44 113 261 7753 neil.crombie@towerswatson.com Alex Browning +44 1737 274254 alex.browning@towerswatson.com Defined benefits pension schemes the impact on FTSE350 company accounts at 31 December 2011 3
Headline results Discount rate Expected future payments from pension schemes are discounted in order to calculate the net present value of the pension liabilities. A lower discount rate means a higher value is placed on the liabilities. The yield on AA rated over 15 year corporate bonds fell by around 0.7% per annum during 2011. This is reflected in the change to the average discount rates adopted by companies, with the average discount rate adopted at 31 December 2011 of 4.8% pa representing a fall of 0.7% pa on that at 31 December 2010. Discount rates could have been up to 0.3% pa higher at 31 December 2011 for those companies using a cashflow-based approach to setting discount rates, when compared to those using an index-based approach. A typical company following the cashflow approach could have disclosed pension liabilities around 5% lower than those using an index approach. Corporate bond yields have fallen by a further 0.3% pa at the time of writing. Inflation Benefits payable from pension schemes are generally linked to inflation. A lower inflation assumption means a lower value is placed on the pension liabilities. Pension schemes can have benefits linked to the Retail Price Index (RPI) or the Consumer Prices Index (CPI). The average RPI assumption adopted by companies fell by around 0.3% pa over the year, representing falls in expected future inflation. The CPI is generally expected to increase more slowly than the RPI in future due to differences between the two indices. Companies generally adopt a CPI assumption by taking a margin from their RPI assumption. The average margin adopted for disclosures at 31 December 2011 was 0.9% pa. Life expectancy The life expectancy assumptions show the length of time that a company is assuming that pensions will be paid to members of its pension scheme. Higher life expectancy assumptions lead to higher pension liabilities. Life expectancy for male pensioners aged 65 ranged from 83.1 to 90.0 years, with the average assumption being 87.3 years. Life expectancy for female pensioners aged 65 ranged from 87.1 to 92.5 years, with the average assumption being 89.3 years. On average, companies assumed that men currently 45 years old would, on reaching 65, live for 2.0 years longer than today s retirees. For women, the average assumption was a further 1.8 years of life expectancy. Expected return on assets The income statement disclosed in company accounts includes a credit for the expected return that a company believes that its pension scheme assets will earn. Companies assuming higher returns from their pension scheme assets will see a lower pension cost in their income statement. Expected return on assets assumptions adopted by FTSE350 companies ranged from 3.4% pa to 7.3% pa, with the average assumption being 5.5% pa. Changes to IAS19 There are some significant changes to IAS19 that will be effective for fiscal years starting on or after 1 January 2013. If these had been in force during 2011, companies reported profits may have been around 4 billion lower. We would expect companies to be considering the impact of these changes in more detail in advance of 31 December 2012. 4 towerswatson.com
The story of 2011 Overview Overall, higher bond prices and additional company contributions meant that companies pension scheme assets were generally worth more at the end of 2011 than they were 12 months earlier, despite equities losing value during the year. However, liabilities have also risen, due mainly to lower interest rates. This has meant that, on average, pension deficits are unchanged from a year earlier. The precise picture varies a lot from company to company depending on how the scheme s assets are invested and how soon its pensions have to be paid. However, most pension scheme sponsors will have experienced another bumpy ride during 2011. We have illustrated in Figure 01 how the combined pension scheme funding for FTSE350 companies has changed over the year. Figure 01. Changes in the aggregate deficit of FTSE350 companies pension schemes since 31 December 2010 Details of improvements in June 90 Equity markets up around 4% 80 Corporate bond yields up around 70 0.1% pa Combined 60 impact = 22 billion reduction 50 in deficit Deficit bn 40 30 20 10 Dec 2010 Mar 2011 Jun 2011 Sep 2011 Dec 2011 Mar 2012 Figure 01 has been produced using Towers Watson s innovative Asset Liability Tracker software. More details in relation to this product are provided in the appendix. The funding position is a measure of how the pension scheme assets compare to the liabilities. A deficit means there is a shortfall of assets relative to liabilities. The figures illustrated in the chart are based on International Accounting Standard 19 (IAS19). Pension liabilities are generally calculated differently for the purpose of pension scheme funding and, as a result, for many companies the pension scheme deficit disclosed in company accounts under IAS19 may be different to the deficit on which future contribution levels are determined. Details of deterioration in October Equity markets down around 5% Corporate bond yields down around 0.3% pa Combined impact = 32 billion increase in deficit Defined benefits pension schemes the impact on FTSE350 company accounts at 31 December 2011 5
Highlights At the start of 2011 the total pension scheme defi cit for FTSE350 companies was around 49 billion. The funding position was extremely volatile during the year. The pension defi cit changed by as much as 32 billion in a single week. To further illustrate the volatility, there were 16 weeks where the pension defi cit for FTSE350 companies changed by more than 10 billion. Falls in equity markets, combined with a fall in corporate bond yields, resulted in the aggregate defi cit rising to around 80 billion in August, its highest point over the year. By October, rising asset values meant that the position was much improved, with a defi cit as low as 4 billion. Towards the end of the year assets did not keep pace with rising pension liabilities and the defi cit had increased back to around 49 billion at 31 December 2011. 2012 has started much the same way as 2011, with a volatile position during the fi rst half of the year. The initial improvement in the defi cit immediately following the year-end was short lived and by 21 May 2012, the defi cit was around 90 billion. In-depth analysis When estimating a scheme s funding position, companies fi rst need to estimate the pension payments that will be paid to members of the scheme. The amount of future pension payments will depend on various unknown factors such as how long people will live and the level of future infl ation. Companies will then discount future payments back to a current date in order to compare the liabilities with the pension scheme assets. The following sections provide a more detailed analysis of the underlying assumptions that impact on pension scheme disclosures in company accounts. It is important to consider the impact of changes to the assumptions together. For example, increases to both the discount rate and infl ation assumption may lead to no signifi cant overall change in the value placed on pension scheme liabilities. We also illustrate in the following sections some key areas for focus in the coming months in the calls to action within each section. Many pension scheme sponsors will have experienced another bumpy ride during 2011. 6 towerswatson.com
Discount rate Of all the assumptions used to estimate a company s pension liabilities, it is the discount rate that has the greatest impact. Pension scheme liabilities refl ect the net present value of expected future payments from the scheme. A lower discount rate means a higher value is placed on each future payment from the scheme, meaning higher liabilities. Accounting standards require that the discount rate is equal to the yield available on high quality corporate bonds, of equivalent maturity and currency to the pension scheme liabilities. The pension liabilities for different companies will have different average terms to maturity refl ecting the age of their membership, with most falling within the range 15 to 25 years. Why were pension liabilities so high at 31 December 2011? Figure 02 shows the movement in the yield available from long-dated corporate bonds over the last fi ve years. The index shows how corporate bond yields have fallen signifi cantly. The yield of 4.7% pa at 31 December 2011 represents a fall of over 0.7% pa since the previous year end. Yields have fallen further in early 2012, and have been at the lowest levels since corporate bond yields became the key reference point for companies calculating the pension liabilities to disclose in their accounts. In May 2012, yields were 4.4% pa. Figure 02. Movement in the yield on long-dated corporate bonds over the last five years Yield (% pa) 8.0 7.5 7.0 6.5 6.7 6.0 5.8 5.7 5.5 5.0 5.4 4.5 4.7 Dec 2007 Dec 2008 Dec 2009 Dec 2010 Dec 2011 Source: iboxx Corporate AA-rated Over 15 Years Index. The impact of a reduction in corporate bonds yields of 1% pa at 31 December 2011 would have been an addition of around 120 billion on FTSE350 pension liabilities. Defi ned benefi ts pension schemes the impact on FTSE350 company accounts at 31 December 2011 7
The impact of UK Quantitative Easing (QE) and the Eurozone debt crisis There have been many press headlines about UK gilt yields in recent months. Many of the headlines have focussed on the impact of low gilt yields on defined benefit pension scheme liabilities and also on the level of individuals retirement income from defined contribution arrangements. Two factors that have contributed to the historically low gilt yields have been the QE policy under which the Bank of England has purchased gilts and the fact that UK gilts are currently seen by investors as a safe haven given the current economic turmoil within the Eurozone. Both of these factors would also indirectly lead to reduction in the yields from corporate bonds. In the case of QE, when the central bank purchases assets, the money holdings of the sellers are increased. Unless money is a perfect substitute for the assets sold, many sellers will have attempted to rebalance their portfolios by buying other assets (such as corporate bonds) that are better substitutes, pushing up prices (and reducing yields). Analysis by the Bank of England suggests that the initial 200 billion of QE may have led to a fall of 1% pa in gilt yields and of 0.7% pa in yields on investment grade corporate bonds around the time of the initial QE announcements. 1 At the time of writing, the total QE programme amounts to around 325 billion. It is important to note that, whilst the QE programme and the Eurozone crisis will undoubtedly have driven down bond yields and led to higher values being placed on pension liabilities, the QE programme will also have increased the value of assets held by pension schemes, thereby reducing the overall impact on pension scheme funding. Why is maturity so important when assessing pension scheme liabilities? Figure 03 shows individual corporate bond yields by maturity at 31 December 2011. The curves show the Towers Watson best fit to the individual yields at both 31 December 2010 and 31 December 2011, calculated using our RATE:Link model. 1 Source: Bank of England Quarterly Bulletin 2011 Q3 Figure 03. Corporate bond yields Yield (% pa) 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 0 5 10 15 20 25 30 35 40 45 50 Maturity (years) 2011 yields 2010 RATE:Link curve 2011 RATE:Link curve Source: Data as published by Bloomberg 8 towerswatson.com
As noted earlier, accounting standards require that the discount rate is equal to the yield available on high quality corporate bonds, of equivalent maturity and currency to the pension scheme liabilities. Typically, companies will set discount rates based on one of two approaches: In line with corporate bond yields as suggested by an index of bonds, taking account of the long-term nature of pension schemes by looking only at an index of long-dated corporate bonds, or; More accurately reflecting the actual duration of their pension scheme by weighting actual corporate bond yields at each maturity by the expected cashflows from the pension scheme. The first approach may not fully reflect the higher yields currently available on very long-dated corporate bonds, as illustrated in Figure 03, as the bonds within the indices are likely to be of shorter maturity than typical pension scheme cashflows (even when only looking at an index of long-dated corporate bonds). Therefore, those setting discount rates based on index yields, as opposed to the cashflow-based approach, may have disclosed higher liabilities at 31 December 2011. For a typical scheme, the impact may be about 5% on liabilities in 2011. It is not possible to identify the approach adopted by each company from their accounting disclosures. The approach selected would have made less difference to the liabilities disclosed at 31 December 2010 because bond yields in the 10 to 30 year range did not increase in the same way. The average reduction in the discount rate adopted by FTSE350 companies over the year was 0.7 % pa the approximate impact of this is an increase in liabilities of around 15%, or around an 80 billion increase in the aggregate FTSE350 deficit. Figure 04. Discount rates adopted by companies at 31 December 2011 Proportion of companies (%) 35 30 25 20 15 Average discount rate 31 December 2010 5.5% Average discount rate 31 December 2011 4.8% 10 5 0 4.4 4.5 4.6 4.7 4.8 4.9 5 5.1 5.2 5.3 5.4 5.5 Percentage per annum Call to action: How would adopting the more accurate cashflow-based approach to discount rate setting affect your pension liabilities? Defined benefits pension schemes the impact on FTSE350 company accounts at 31 December 2011 9
Inflation Future price inflation is an important assumption as benefits payable from pension schemes are generally linked to inflation. Lower price inflation means a lower value is placed on the scheme s liabilities. In the past, companies have only been concerned about the rate at which the RPI increases. However, a lot of pension schemes now have benefits linked to the CPI following the Government s decision to use CPI as the as the basis for the statutory minimum revaluation and indexation for pensions provided by occupational pension schemes. RPI The approach taken by companies has typically been to adopt an RPI assumption by looking at the inflation implied by gilt prices (with implied inflation reflecting the difference in yields from fixed interest and index-linked gilts). There are two main reasons why different companies will disclose different assumptions about RPI inflation. First, they all estimate average inflation rates over different time periods, corresponding to when inflation-based pension increases will be awarded in their particular scheme. In general, market implied inflation at 31 December 2011 was higher at dates further in the future, so it could be expected that more mature schemes might use lower inflation assumptions. Second, companies may have reached different judgements about the extent to which the inflation rates implied by gilt prices reflect expectations of what inflation will actually be as opposed to the premium that pension funds and insurance companies will pay for inflation-linked assets that match their liabilities. Changes in the last five years Figure 05 illustrates how market expectations for future RPI inflation have moved over the last five years. This shows how long-term inflation expectations have not been as volatile as changes to discount rates. Figure 05. Changes in market expectations of RPI inflation over the last five years 5.0 8.0 4.5 7.5 4.0 3.8 3.8 7.0 3.5 3.5 3.3 6.5 3.0 3.2 6.0 2.5 5.5 2.0 5.0 1.5 4.5 1.0 4.0 Dec 2007 Dec 2008 Dec 2009 Dec 2010 Dec 2011 RPI inflation (% per annum) Corporate bond yields (% per annum) The inflation in the chart is that implied by the gilt pricing, with an average payment date of 20 years in the future. 10 towerswatson.com
Assumptions adopted by companies at 31 December 2011 In Figure 06 we show how expected future inflation, as implied by gilt prices, has fallen by around 0.5% pa at longer durations with a slightly lower fall at shorter durations. Figure 06. Expected future inflation implied by gilt prices % per annum 4.00 3.75 3.50 3.25 3.00 2.75 2.50 2.25 2.00 0 5 10 15 20 25 Term (years) 31 December 2011 31 December 2010 The actual RPI assumptions adopted by FTSE350 companies are set out in Figure 07. Figure 07. RPI assumptions adopted at 31 December 2011 Proportion of companies (%) 40 35 30 25 20 15 Average RPI assumption at 31 December 2010 3.4% pa Average RPI assumption at 31 December 2011 3.1% pa 10 5 0 2.5 2.6 2.7 2.8 2.9 3.0 3.1 3.2 3.3 3.4 3.5 Percentage per annum A reduction in expected future inflation assumption of 1% pa would result in a reduction in pension liabilities of around 17%, equating to around 100 billion for FTSE350 companies. Defined benefits pension schemes the impact on FTSE350 company accounts at 31 December 2011 11
What is the difference between RPI and CPI? There are two key areas where RPI and CPI differ: The calculation methodology. The composition of the underlying basket of goods and services (there are many differences, one of the key differences being the inclusion of council tax and owner occupiers housing costs in RPI but not, currently, in CPI). The Office for Budget Responsibility published a report in November 2011 with an estimate of the long-term gap between RPI and CPI of 1.3% to 1.5% pa (with CPI inflation expected to be lower than RPI inflation). CPI The assumption adopted for CPI inflation is more subjective due to the lack of a liquid market in CPI priced gilts. Figure 08 illustrates the range of assumptions adopted by FTSE350 companies for the long-term gap between RPI and CPI inflation at 31 December 2011. The average assumed gap is 0.9% pa and none of the companies whose results are included in this report have used the Office for Budget Responsibility s assumption that it will be 1.4% pa. Accounting for the change to CPI Typically, the pension scheme liabilities of FTSE350 companies will have reduced following the Government s decision to use CPI as the basis for statutory minimum pension increases. However, the magnitude of this reduction will vary from scheme to scheme. The two main areas where companies could see savings are on the uprating of pensions in payment and on the revaluation of pensions for members who no longer accrue benefits in the pension scheme, but who have not yet retired. Either of these could have changed following the Government s decision. The impact will depend on the drafting of individual scheme rules. At the time of reporting for 31 December 2010, the precise legal position for some schemes was not clear. As a result, some companies made an allowance for anticipated savings in their 2010 year end disclosures whilst others chose to defer making an allowance until the legal position was clarified (allowance is likely to have been made in the 2011 year end disclosures). However, the size of the liability reduction specifically attributed to the CPI change was often not disclosed. For the majority of companies, the expected savings for moving from RPI to CPI-based pension increases will have been recognised in their accounts through OCI (Other Comprehensive Income), reflecting a change in the assumption for future pension increases. However, for those companies who felt that members previously had valid expectations of RPI increases (perhaps due to a reference to RPI in scheme literature), allowance for the anticipated saving from moving to CPI could have been recognised in their income statement (commonly referred to as the profit and loss account), boosting their profitability for the year. Call to action: Consider the margin adopted for CPI relative to RPI and the impact of any minimum and maximum rules that apply to pension increases in your scheme. Figure 08. Assumption for long-term gap between RPI and CPI at 31 December 2011 Proportion of companies (%) 40 35 30 25 20 15 Average CPI to RPI margin at 31 December 2011 0.9% pa 10 5 0 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 Percentage per annum 12 towerswatson.com
Life expectancy The life expectancy assumptions adopted by companies show the length of time that a company is assuming that pensions will be paid to members of its pension scheme. Higher life expectancy assumptions will lead to higher pension liabilities. Companies tend to disclose two life expectancy assumptions; that of people retiring now and also that for pension scheme members who will retire in future. This provides the users of accounts with an indication of the level of future improvements in life expectancy that the company is allowing for within their accounting disclosures. Current retirees The starting point for many companies in setting the current life expectancy will be the latest study by the Continuous Mortality Investigation Bureau into the life expectancy of Self-Administered Pension Scheme (SAPS) members. A scheme-specific adjustment may then be applied, either based on a scheme s own observed mortality experience, or using an indirect method such as postcode analysis which aims to reflect the key lifestyle, health and wealth characteristics of the scheme s membership. The life expectancy of current 65 year-olds will also reflect expectations of longevity improvements over the remaining lifetimes of those individuals, which can vary from company to company. When taken together, the starting point and expectation of improvements can lead to significant divergence in life expectancies between different pension schemes. Figure 09. Assumed life expectancy for members aged 65 at 31 December 2011 Percentage of companies 50 40 30 20 10 0 83-84 Male 84-85 85-86 Female 86-87 87-88 88-89 The average life expectancy assumption for male pensioners aged 65 at 31 December 2011 was 87.3 years. The average life expectancy assumption for female pensioners aged 65 at 31 December 2011 was 89.3 years. 89-90 90-91 Around 80% of companies reporting at 31 December 2011 adopted an assumption in line with the SAPS study (based on a Towers Watson survey). 91-92 92-93 Age range Defined benefits pension schemes the impact on FTSE350 company accounts at 31 December 2011 13
Future retirees Comparing the life expectancy assumed for future retirees to that assumed for current retirees allows us to consider the allowance for future improvements in isolation. The level of future improvements in life expectancy is subject to a wide range of opinion, but generally will be informed by a combination of past observation and cause and disease-based modelling. There are various tools, such as the Towers Watson Stochastic Mortality Risk Model, that can help companies to determine a central estimate for the future long-term rate of mortality improvement, as well as helping to illustrate the potential risk due to the uncertainty of future life expectancy. The allowance for future improvements in life expectancy for an individual aged 45 on reaching age 65 (relative to a current 65 year old) is shown in Figure 10. On average, companies assumed that current 45 year-old men would, on reaching 65, live for 2.0 years longer than today s retirees. For women, the average was a further 1.8 years of life expectancy. Changes since 31 December 2010 If we compare the average life expectancies at 31 December 2011 with those at 31 December 2010, we can see that life expectancy assumptions have not changed significantly over the year. How do the assumptions compare to current population statistics? By comparison, the latest Offi ce for National Statistics projections assume that, for the UK population as a whole: Average life expectancy for people aged 65 in 2011 is 86.1 for men and 88.8 for women. These are lower than the life expectancies assumed by FTSE350 companies as pension scheme members have historically lived longer than the population as a whole. For current 45 year olds, on reaching age 65, average life expectancy is projected to be 88.6 years for men and 91.1 for women. Call to action: This section illustrates potential variability in life expectancies. Is now a good time to consider hedging your exposure to changes in future life expectancies? Figure 10. Assumed level of future improvement in life expectancy for members age 45 on reaching 65, relative to current 65 year olds Percentage of companies 50 40 30 20 10 0 15 0-0.25 35 0.5-0.75 0.25-0.5 3.5-3.75 3.25-3.5 3-3.25 2.75-3 2.5-2.75 2.25-2.5 2-2.25 1.75-2 1.5-1.75 1.25-1.5 1-1.25 0.75-1 15 23 8 33 19 4 5 5 Male Female Additional life expectancy (years) 14 towerswatson.com An increase in life expectancy of one year would add around 3%, or approximately 20 billion, to FTSE350 pension liabilities.
Expected return on assets The assumptions we have discussed so far in this report are used to place a value on companies pension liabilities. However, the income statement disclosed in company accounts includes a credit for the expected return that a company believes that its pension scheme assets will earn. Companies assuming higher returns from their pension scheme assets will see a lower pension cost in their income statement, therefore boosting profi tability. The expected return on asset assumption is determined by the mix of assets held within the pension scheme and by companies views on the expected returns from various asset classes. A simple example of the expected return on asset assumption is given in Figure 14. Mix of assets In recent years there has been a general move to reduce the level of risk within pension schemes. One of actions taken to reduce risk has been to sell return-seeking investments, such as equities, in favour of matching assets, such as bonds. Figure 11 illustrates how the mix of pension scheme investments changed in aggregate over 2011, some of which will be as a result of changes in the relative prices of the various asset classes and some of which will be due to pension scheme de-risking. Figure 11. Pension scheme investments at 31 December 2010 and 31 December 2011 4% 47% 11% 2011 ( 530bn) 2010 ( 490bn) 10% 4% 43% 43% 38% Equities Bonds Property Other A reduction in the expected return on assets assumption of 0.5% pa would increase the pensions charge for FTSE350 companies in 2012 by around 3 billion. One consequence of moving towards a lower risk investment strategy is a reduction in the expected returns from the pension scheme assets. This leads to a reduction in the credit disclosed in income statements and, therefore, a higher pension cost for those companies. Defi ned benefi ts pension schemes the impact on FTSE350 company accounts at 31 December 2011 15
Expected returns from different asset classes The expected returns from some asset classes, such as bonds, can be derived from market yields. The expected returns from other assets, such as equities, will be derived using a combination of modelling and judgement, and can vary signifi cantly from company to company. A combination of the different assets held, together with the subjective expected returns from each asset class, can lead to signifi cant differences in the overall level of returns assumed by different companies. Figure 12 shows the range of assumptions adopted by companies at 31 December 2010 and 31 December 2011. The median assumption has reduced over the year by 0.9% pa. Figure 12. Expected returns on pension scheme assets Median 2011 Minimum Maximum Lower quartile Upper quartile 2010 Minimum Maximum Median 3 4 5 6 7 8 9 Expected return on asset assumption (% per annum) Call to action: How will the changes to IAS19 effective from 1 January 2013, described in the next section, impact on the expected return on assets component of your income statement? 16 towerswatson.com
Future changes to IAS19 In June 2011, the IASB issued an amendment to IAS19 (Employee Benefits) that will come into force for fiscal years starting on or after 1 January 2013. In this section we focus on two of the main changes. The key changes for pension scheme sponsors to consider are: Expected return on assets: Instead of crediting asset returns through the income statement based on the expected return on the pension scheme assets, the credit on assets will be the same as the discount rate, potentially reducing the stated profi ts of the majority of companies. Immediate balance sheet recognition: Gains and losses arising from experience differing from that assumed, changes in assumptions, investment gains and losses on assets, and the effect of pension scheme changes will need to be immediately recognised. Expected return on assets The change to the expected return on pension scheme assets which is included within the income statement is likely to affect the majority of FTSE350 companies. Instead of crediting asset returns through the income statement based on the expected return from the pension scheme s assets, the credit on assets will be based on the discount rate used to value the liabilities. The impact of this change will vary for each company and is highly dependant both on the asset allocation and on each company s current assumption for future asset returns. The difference between the expected return on asset assumption and the discount rate adopted by each company at 31 December 2011 is illustrated in Figure 13. Those with the largest positive difference between the expected return on asset assumption and the discount rate will see the biggest hit to profi tability (those with a negative difference will see profi tability improve). This change removes an accounting advantage associated with taking investment risk. However, it does not affect companies main motivation for doing this the hope that higher returns will reduce the cash contributions needed to pay off deficits. Figure 13. Difference between expected return on assets and discount rates at 31 December 2011-1.5-1.0-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Difference (% pa) Average expected return on asset assumption at 31 December 2011 = 5.5% pa Average discount rate at 31 December 2011 = 4.8% pa Defi ned benefi ts pension schemes the impact on FTSE350 company accounts at 31 December 2011 17
If the change had been implemented as at 31 December 2011, based on the average difference between the expected return on assets and discount rate of 0.7% pa, we estimate that the reported profits of the combined FTSE350 would reduce by around 4 billion. The impact of the change to expected return on assets is illustrated in the simple example shown in Figure 14. You will see that the pension cost for this company has increased, reflecting the reduced credit for the expected return from pension scheme assets from 2013 onwards. In addition to the above impact, companies will need to consider the treatment of investment management expenses. Currently, the expected return on assets assumption should be net of any investment management expenses, effectively meaning such expenses are included as a cost within a company s income statement. However, under the revised IAS19 investment management expenses will need to be included in Other Comprehensive Income. Figure 14. Expected returns on pension scheme assets 2011 2012 2011 recalculated on 2013 basis 2012 recalculated on 2013 basis Discount rate 5.5% pa 4.8% pa 5.5% pa 4.8% pa Expected return 6.2% pa 5.7% pa Same as discount rate Same as discount rate on assets Assets 450m 475m 450m 475m Service cost 8m 8m 8.0m 8m Interest cost 27m 28m 27m 28m Expected return on assets ( 28m) ( 27m) ( 25m) ( 23m) Total pension cost 7m 9m 10m 13m 18 towerswatson.com
Immediate balance sheet recognition Currently, companies have two choices within their accounts as to how to recognise gains or losses on their pension obligations: Immediate recognition through Other Comprehensive Income. Using a corridor approach to spread gains and losses, with any gains or losses recognised flowing through the company s income statement. This second option will no longer be available and, as a result, immediate balance sheet recognition will need to be adopted for all FTSE350 companies reporting under IAS19. Around 10% of FTSE350 companies currently use the corridor approach. For those companies who reported at 31 December 2011 and adopted the corridor approach, had they accounted under the revised IAS19 their balance sheet position would have deteriorated by around 30 billion. We would expect those companies that might be affected by these changes to be considering this in more detail in advance of 31 December 2012. IAS19 is changing in 2013. Make sure you are up to speed with the required changes. Call to action: How will you communicate the impact of these changes to investors? Defined benefits pension schemes the impact on FTSE350 company accounts at 31 December 2011 19
Appendix How Towers Watson can help pension scheme sponsors We have illustrated in this section some of the Tower Watson software we used within the main body of this report. This software has been designed to help sponsors and trustees to best manage their pension obligations. Asset Liability Tracker Asset Liability Tracker is an online system that provides a platform to enable sponsors and trustees to manage their pension schemes in dynamic and rapidly changing markets. It provides daily updates of the pension scheme s position and the status of any trigger points for action (for example triggers for investment strategy changes). This enables sponsors and trustees to act quickly and optimise use of the scheme s assets and sponsors contributions. We have used this software to track the funding position of the FTSE350 pension schemes in The story of 2011 section of this report. Figure 15. Output from Asset Liability Tracker 20 towerswatson.com
Global RATE:Link Towers Watson s Global RATE:Link is a tool designed to assist sponsors in the selection of discount rates that accurately refl ect the characteristics of their pension schemes. Figure 16. RATE:Link output for the UK at 31 December 2011 The Global RATE:Link model calculates hypothetical yield curves developed from corporate bond yield information within each regional market. The yields are derived from bonds across a range of maturity points, and a curve is fi tted to those targets. The present value of plan benefi ts (the value of the liabilities) is calculated by applying the discount rates to expected benefi t cashfl ows. A single discount rate is then developed to produce that same present value this represents the suggested discount rate. Figure 17. Illustrating the potential impact of life expectancy on pension scheme cashflows This software has been used to illustrate the higher yields available from longer-dated corporate bonds in the Discount Rate section of the report. Stochastic Mortality Risk Model The amounts that a pension scheme needs to pay its members depend on two key factors, the payments promised and whether a member is alive to receive the payment. To assess the level of cash payments required and by how much they can vary, it is important to understand the likely life expectancy (longevity) of the scheme s membership, and the fi nancial impact of longevity being different to expectations. Towers Watson s Stochastic Mortality Risk Model is an innovative tool which provides quantitative analysis to help assess the longevity risk associated with a scheme s population. Using the model, we can model two key elements of the mortality risk associated with a scheme s population: Figure 18. Prudence within life expectancy assumptions Trend risk the risk that future mortality improvements affecting a scheme, and hence underlying future lifespans, are different from the central assumption. Idiosyncratic risk the risk that, even if mortality rates now and in future are known with certainty, by chance particular individuals or groups live longer than expected. Figure 18 shows how the software can also be used in funding negotiations to illustrate the level of prudence within current life expectancy assumptions. Defi ned benefi ts pension schemes the impact on FTSE350 company accounts at 31 December 2011 21
22 towerswatson.com
Defined benefits pension schemes the impact on FTSE350 company accounts at 31 December 2011 23
About Towers Watson Towers Watson is a leading global professional services company that helps organisations improve performance through effective people, risk and financial management. With 14,000 associates around the world, we offer solutions in the areas of employee benefits, talent management, rewards, and risk and capital management. Towers Watson 21 Tothill Street Westminster London SW1H 9LL Towers Watson is represented in the UK by Towers Watson Limited and Towers Watson Capital Markets Limited. The information in this publication is of general interest and guidance. Action should not be taken on the basis of any article without seeking specific advice. To unsubscribe, email eu.unsubscribe@towerswatson.com with the publication name as the subject and include your name, title and company address. Copyright 2012 Towers Watson. All rights reserved. TW-EU-2012-24283. June 2012. towerswatson.com