Global Report: Global Survey of Retirement Plan … · Global Report: Global Survey of Retirement Plan Accounting Assumptions . ... Canadian AA bond universeand created challenges

Aon Hewitt Legislative Reporting

Risk. Reinsurance. Human Resources.

Global Report: Global Survey of Retirement Plan Accounting Assumptions This report presents the results of Aon Hewitt’s global survey of accounting assumptions used for employee benefit plans at 2016 year-end. The results of this survey may be useful to companies when setting preliminary assumptions for 2017 year-end and budgets for 2018. In addition, companies should consider the following:

At 2016 year-end, discount rates in most countries decreased significantly once again from the prior year-end. The yields on corporate bonds in countries with large pension liabilities (i.e., Canada, the Eurozone, Switzerland, the United Kingdom, and the United States) have subsequently moved in different directions. During the first half of 2017, rates have decreased significantly since 2016 year-end in the U.S. and Canada, have increased in the Eurozone, and have remained relatively stable in Switzerland and the U.K. Any planning for 2017 year-end or budgeting for 2018 should take these and subsequent movements into consideration.

Changes in bond yields mentioned above may be due, in part, to changes in inflationary expectations. When plan sponsors consider a change in their discount rate, they should review other assumptions linked to inflation.

The average expected rate of return on plan assets in many countries continues to decline due to a combination of increased allocations to fixed-income investments and reduced future outlooks for long-term returns on some asset classes. For plan sponsors using International Financial Reporting Standards (IFRS), this decline will not affect their pension expense because the revised version of IAS 19, effective for fiscal years beginning in 2013, does not call for this assumption to be used.

In Canada, a GE Capital bond repurchase in the first half of 2016 materially impacted yields at the long end of the Canadian AA bond universe and created challenges for discount rate curve construction due to sparse data. Discount rates derived from the PC Bond yield curve also fell dramatically, and this generated questions on the efficacy of this methodology.

We are seeing companies, especially those with closed or frozen plans, establish a “glide path” as part of their investment and risk policy, with a view to automatically adjusting the risk level in the plans as market conditions vary. The glide path typically sets a specific asset allocation the investment manager must follow based on predefined triggers such as the funded status of the pension plan. If such a policy is applied, plan sponsors using U.S. GAAP should consider how the asset allocation in the relevant plans may vary when setting an assumption for the expected rate of return on assets.

July 2017

Global Survey Retirement Plan Accounting Assumptions—YE2016 | Aon Hewitt | July 2017 2

Most countries in which defined benefit plans are prevalent have mortality assumptions that recognize anticipated future mortality improvements. Below are comments concerning mortality assumptions in several countries:

– United Kingdom—The Continuous Mortality Investigation model was reviewed in 2016, resulting in new projection scales being released in March 2017. This new scale again showed a reduction in projected future mortality improvements, which will generally result in a decrease to obligations and service costs. Some companies used data which was publicly available in 2016 to generate a “proposed” projection scale prior to the final scales being released. The updated projection methodology may result in less pronounced mortality adjustments than during the past two years.

– United States—In October 2014, new mortality assumptions were issued in the United States. Many plan sponsors adopted these new assumptions for 2014 year-end. Some clients performed mortality studies and adjusted the standard tables to better fit their experience. Since 2014, additional data has shown a slowdown in mortality improvements for the general population in the U.S., and new mortality improvement scales reflecting this experience have been issued in 2015 and 2016. These updated improvement scales have been adopted by many companies.

– Switzerland —Alternate mortality improvement projections based on the Continuous Mortality Investigation Model, recently developed in U.K. and mentioned above, are under investigation. Based on recent feedback from auditors, it appears that such an alternative approach is acceptable as a best estimate assumption.

Companies have been using additional ways to reduce the financial risks related to their pension plans, such as lump-sum windows, buy-ins, and longevity swaps. Lump-sum windows are typically temporary and offer members an opportunity to take a lump-sum payment in lieu of an annuity. Buy-ins occur when the pension plan purchases an annuity contract as an investment but retains the ultimate obligation to pay benefits. Under IFRS, buy-ins are generally treated as investment decisions and not as settlements, with the fair value of the annuity contract set equal to the related Defined Benefit Obligation. While the treatment is less clear under U.S. GAAP, we have, in general, not seen buy-ins trigger settlement accounting under U.S. GAAP either; however, the measurement of the annuity contract will generally differ from the IAS 19 approach, with the Projected Benefit Obligation set equal to the fair value of the annuity contract. Longevity swaps are contracts between the pension plan and an insurance company under which the insurance company agrees to take on the risk related to retirees living longer than is currently expected.

An increasing number of companies are now using a yield curve approach (also known as the “spot rate” approach) to calculate the service cost and interest cost for their pension and other postretirement plans. In the past, the single equivalent discount rate used to calculate the benefit obligation was typically used to calculate these values. At this point, approximately 50 percent of companies which publicly report under U.S. GAAP have adopted this spot rate approach. For companies using mark-to-market pension accounting, the spot rate approach does not change the total pension expense for the year. It just changes how and when the expense is reported. Companies not using mark-to-market pension accounting will have their annual expense change and will need to address this with their auditors if they wish to adopt this method.

A significantly smaller number of companies which report under IFRS have adopted the spot rate approach so far:

– An additional complication when applying the approach under IFRS is that the interest cost is calculated on the net pension liability (i.e., pension liability less pension assets). Using the spot rate approach requires separately calculating the interest cost for liabilities and assets. While it is clear how to calculate the interest cost for the liabilities, it is not clear how the offsetting interest benefit for the assets is calculated. One approach used by the companies that have adopted under IFRS is to use the “average” discount rate used for the liabilities.

– Perhaps partly because the approach to determining interest on the net pension liability has not yet been widely adopted, some companies have adopted the spot rate approach only for the service cost calculation.

– In any event, for companies that report under IFRS, the impact on pension expense of moving to the approach is generally less significant than adopting the approach under US GAAP because the interest credit on the assets is impacted in exactly the same way as the interest cost on the liabilities.

Although we have observed reluctance from some audit firms, we expect that most auditors will accept using the spot rate approach to determine both the service cost and interest cost, similar to U.S. GAAP. However, since this approach is not prevalent under IFRS, clients should expect to spend time discussing the merits of the spot rate approach with their auditors.

In March 2017, the U.S. Financial Accounting Standards Board issued Accounting Standards Update (ASU) 2017-07, which will impact companies reporting under U.S. GAAP. ASU 2017-07 requires that the service cost component of pension expense be presented in the same financial statement line item(s) as other employee compensation costs, and


that the other components of expense be presented separately and outside of any subtotal of income from operations. In addition, ASU 2017-07 allows only the service cost component of expense to be capitalized (e.g., as a cost of inventory or a self-constructed asset). ASU 2017-07 is effective for public business entities for annual periods beginning after December 15, 2017, including interim periods within those annual periods. A delayed effective date applies for other entities. Early adoption is permitted. The changes in financial statement presentation requirements must be applied retrospectively, but the changes in capitalization requirements only apply prospectively.

Proposals in the IASB 2015 Exposure Draft are still not yet final. The following changes to IAS 19 were proposed:

– Clarification of the application of IFRIC 14’s limit on a pension asset in specific situations;

– Clarification of the treatment of the IFRIC 14 asset ceiling in the event of an amendment, curtailment, or settlement; and

– Revising IAS 19 so that the expense for periods after an amendment, curtailment, or settlement is recognized to reflect those events similar to U.S. GAAP.

Since the IASB's decision to adopt the proposed changes in 2016, some minor wording changes have been deliberated which could potentially mean IFRIC 14 would restrict surplus recognition for a much larger proportion of UK companies. Various parties have raised their concern with the IASB and clarification is expected before the end of 2017. We believe any changes to IAS 19 are likely to come into force for years beginning on or after 1 January 2019 at the earliest.

Due to an ultimate employer liability, every pension promise in Germany technically qualifies as a defined benefit plan under the main accounting standards. However, in line with IAS 19.46, it is common market practice, and generally accepted by auditors, to treat certain insured plans as defined contribution plans. As the ongoing low interest rate environment also affects insured plans, a review of the accounting treatment of such plans may be necessary in some cases. The German Institute of Auditors (Institut der Wirtschaftsprüfer – IDW) has therefore issued guidance on how to treat a change from defined contribution accounting to defined benefit accounting. If such a change is based on a new economic assessment of the overall plan rules, the effects are recognized via Other Comprehensive Income; if, on the other hand, the change in accounting is resulting from a change in an accounting policy, the effect should be recognized through the P&L, in line with IAS 8.

If you would like an update on current economic indices or the current status of changes in accounting standards, please contact your local Aon Hewitt consultant.


Background Aon Hewitt conducted this survey of fiscal 2016 year-end accounting assumptions by gathering assumptions used by our clients. Previous surveys also included results for companies that may have used other actuarial firms.

As in previous surveys, we have focused on four economic assumptions that companies select under ASC 715, IAS 19, FRS 102 and other accounting standards with similar requirements for assumption setting. The assumptions are:

Discount rate;

Expected long-term rate of return (U.S. GAAP only);

Salary increase; and

Pension increase assumptions (for certain countries only)

Similar to last year, we are only showing assumptions for companies with fiscal years ending on December 31, 2016. The tables on the following pages show the average assumption as of the end of 2015 and 2016.

The Appendix provides the following additional information about 2016 year-end assumptions:

Average discount rate by duration of the pension liabilities for countries with larger pension obligations (Canada, Eurozone, Japan, Switzerland, and the U.K.);

Range of assumptions used at 2016 year-end for specific countries; and

Prevalent mortality table along with life expectancies for countries with larger pension liabilities.


Discount Rate at Year-End 2016 The following table shows the survey results for the discount rate assumption for the 2015 and 2016 fiscal year-ends.

Country Year-End

2015 Average Year-End

2016 Average

Country Year-End


2016 Average Australia 4.02% 3.79% India 7.89% 6.71% Brazil 13.97 12.09 Japan 0.86 0.59 Canada 4.05 3.91 Mexico 6.61 7.46 Eurozone Philippines 4.40 4.75 Austria 2.07 1.46 Poland 2.99 3.39 Belgium 1.88 1.36 South Africa 9.78 9.50 Finland 1.99 1.46 South Korea 2.78 2.66 France 2.07 1.41 Sweden 2.69 2.23 Germany 2.39 1.71 Switzerland 0.78 0.52 Greece 2.21 1.47 Taiwan 1.71 1.39 Ireland 2.55 1.80 Thailand 3.04 2.89 Italy 2.07 1.40 United Kingdom 3.77 2.71 Netherlands 2.34 1.72 United States 4.39 4.17 Spain 2.27 1.57

The values above represent an average discount rate for all post-employment plans (pension, retiree medical, and termination indemnities), as well as in-service benefits such as long-service leave or jubilee awards.

A comparison of a company’s discount rate to the rates shown above should take into consideration: 1) the maturity of the company’s plan(s) may differ from the maturities of the plans included in this survey; and 2) some clients use yield curves which may support a higher discount rate while others may only refer to yields of indices.

See the Appendix for additional details about discount rates for selected countries.


Long-Term Rate of Return for 2017 The following table shows the survey results for the average expected long-term rate of return on plan assets used to determine expense for the 2016 and 2017 fiscal years for companies reporting using U.S. GAAP.

Country 2016

Average 2017

Average

Country 2016

Average 2017

Average Canada 5.16% 5.14% South Korea 3.31% 2.20% Eurozone Switzerland 2.54 2.17 Belgium 3.98 3.81 Taiwan 2.04 1.87 Germany 3.54 2.91 United Kingdom 5.89 5.15 Ireland 3.80 3.56 United States 6.59 6.34 Netherlands 2.43 2.12

As noted on page 1, the concept of an expected return on plan assets no longer exists under IAS 19—it is set equal to the discount rate. For that reason, in this survey, we only included the expected return for companies reporting under U.S. GAAP. As a consequence, the number of countries for which we had credible data declined, and the results for those countries with insufficient data (i.e., those with less than five data points) were removed from this part of the survey.

The expected rates of return shown above for all countries aside from the UK reflect the asset allocations of the plans included in this survey. A comparison of your company’s expected rate of return to the rates shown above should take into consideration that the asset allocation for your company’s plan(s) may differ from the asset allocations of the plans included in this survey. For the United Kingdom, our survey shows the average expected return for equities.

See the Appendix for additional details about the expected long-term rate of return assumption for selected countries.


Salary Increase at Year-End 2016 The following table shows the survey results for the salary increase assumption for 2015 and 2016 fiscal year-ends.

Country Year-End


2016 Average

Country Year-End


2016 Average Australia 3.21% 3.31% India 9.01% 8.72% Canada 3.12 3.08 Japan 2.15 2.19 Eurozone Mexico 5.29 5.18 Austria 2.67 2.60 Philippines 5.46 5.66 Belgium 2.93 2.78 Poland 2.72 2.85 Finland 2.57 2.25 South Korea 4.66 4.47 France 2.47 2.50 Sweden 2.69 2.67 Germany 2.58 2.54 Switzerland 1.52 1.41 Greece 2.06 2.00 Taiwan 3.28 3.22 Ireland 2.48 2.41 Thailand 4.94 5.27 Italy 2.78 2.51 United Kingdom 3.08 3.30 Netherlands 2.24 2.07 United States 3.71 3.79 Spain 2.33 2.20

The salary increase assumptions shown above reflect the situation of each company included in this survey. Your company’s situation may differ from that of the companies included in the survey; hence, a different salary increase assumption may be appropriate.

See the Appendix for additional details about the salary increase assumption for selected countries.


Pension Increase for Year-End 2016 The following table shows the survey results for the pension increase assumption for 2015 and 2016 fiscal year-ends. Note that the results are shown for those plans where a portion of the benefits paid to pensioners is adjusted annually and is correlated with current inflationary expectations.

Country Year-End


2016 Average

Eurozone Germany 1.74% 1.67% Ireland 1.68 1.73 Netherlands 0.91 0.85 Sweden 1.51 1.48 Switzerland 0.04 0.03 United Kingdom 3.14/2.08 3.31/2.25

For the United Kingdom, our 2016 year-end survey results produced an average assumption of 3.31% for the Retail Price Index (RPI) and 2.25% for the Consumer Price Index (CPI).

For Switzerland, we have shown the average pension indexation assumptions for all plans which call for pension indexation, including those with an assumption of 0%.

See the Appendix for additional details about the pension increase assumption for selected countries.


Appendix The Appendix provides the following additional information about 2016 year-end assumptions:

For countries with larger pension obligations (Canada, Eurozone, Japan, Switzerland, and the U.K.), the average discount rate by duration. Note: we were only able to include plans for which the duration of the plan was available;

For specific countries, the range of assumptions used at 2016 year-end; and

For certain countries, the prevalent mortality table used at 2016 year-end along with life expectancies.


The following charts show the average discount rate selected for plans with durations falling within the ranges shown. We also show the number of plans falling within each range. Please note that duration was not provided for all plans included in the survey. Therefore the number of plans included in the charts below may differ from the number of plans used in other parts of this survey.

Canada

Eurozone


Switzerland

Japan


U.K.


The following charts show range of assumptions employers selected for 2016 year-end for specific countries.

Statistic Discount Rate Expected Return Salary ScalePension

Indexation

Count 97 33 76 10Average 3.91% 5.14% 3.08% 1.73%90th %ile 4.27% 6.48% 3.50% 2.10%75th %ile 3.87% 5.80% 3.50% 2.00%Median 3.80% 5.25% 3.00% 2.00%25th %ile 3.71% 4.65% 3.00% 1.39%10th %ile 3.50% 3.55% 2.50% 0.93%

Legend

- Average75th-90th %ile25th-75th %ile10th-25th %ile

0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

6.00%

7.00%

8.00%

DiscountRate

ExpectedReturn

Salary Scale PensionIndexation

Canada


Statistic Discount Rate Salary ScalePension

Indexation Inflation


Legend


0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

3.50%

4.00%

4.50%

DiscountRate


Inflation

France



Indexation


Legend


0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

6.00%

7.00%

DiscountRate

ExpectedReturn


Germany


Statistic Discount Rate Expected Return Salary Scale

Count 39 7 31Average 0.59% 2.73% 2.19%90th %ile 0.80% 4.20% 3.00%75th %ile 0.60% 3.60% 2.50%Median 0.50% 3.10% 2.00%25th %ile 0.40% 1.65% 2.00%10th %ile 0.30% 1.00% 1.50%

Legend


0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

3.50%

4.00%

4.50%

5.00%

DiscountRate

ExpectedReturn

Salary Scale

Japan



Indexation


Legend


0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

6.00%

DiscountRate

ExpectedReturn


Netherlands



Indexation Inflation

Count 153 37 150 150 149Average 0.52% 2.17% 1.41% 0.03% 1.00%90th %ile 0.65% 3.22% 2.00% 0.00% 1.20%75th %ile 0.60% 2.50% 1.50% 0.00% 1.00%Median 0.60% 2.15% 1.50% 0.00% 1.00%25th %ile 0.48% 1.50% 1.00% 0.00% 1.00%10th %ile 0.26% 1.12% 1.00% 0.00% 0.85%

Legend


0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

6.00%

DiscountRate

ExpectedReturn

SalaryScale

PensionIndexation

Inflation

Switzerland


Statistic Discount Rate Expected Return Salary Scale RPI Inflation RPI-CPI Spread

Count 118 30 68 116 114Average 2.71% 5.15% 3.30% 3.31% 1.06%90th %ile 2.80% 6.50% 4.25% 3.50% 1.10%75th %ile 2.75% 6.43% 3.80% 3.40% 1.10%Median 2.70% 5.53% 3.38% 3.30% 1.10%25th %ile 2.65% 4.25% 2.98% 3.20% 1.10%10th %ile 2.60% 2.79% 2.20% 3.20% 1.00%

Legend


0.00%

1.00%

2.00%

3.00%

4.00%

5.00%

6.00%

7.00%

8.00%

DiscountRate

ExpectedReturn

SalaryScale

RPI Inflation RPI-CPISpread

United Kingdom


Statistic Discount Rate Expected Returns Salary Scale

Count 268 226 133Average 4.17% 6.34% 3.79%90th %ile 4.42% 7.50% 4.99%75th %ile 4.30% 7.00% 4.30%Median 4.20% 6.50% 3.75%25th %ile 4.03% 5.75% 3.38%10th %ile 3.90% 5.00% 3.00%

Legend- Average75th-90th %ile25th-75th %ile10th-25th %ile

0.00%1.00%2.00%3.00%4.00%5.00%6.00%7.00%8.00%9.00%

Discount Rate Expected Returns Salary Scale

U.S.


Below is a table showing the prevalent mortality table used for 2016 year-end for the countries shown, along with the life expectancies for a male and female retiree age 65. Please note that we show the life expectancy for a retiree age 65 in 2017 as well as for a retiree age 65 in 2057. The difference between the two values shows the increase in longevity that is expected over the next 40 years by the mortality improvement assumption. For example, an age 65 Canadian male’s longevity is expected to increase by 2.1 years over the next 40 years.

Base Table Life Expectancy (Mortality Improvement if Male In: Female In: Country Separate From Base Table) 2017 2057 2017 2057 Canada CPM Private (Improvement Scale B) 21.6 23.7 24.1 25.9 Germany Heubeck Richttafeln RT 2005 G 19.3 24.3 23.3 28.1 Netherlands Prognosetafel 2016 20.1 24.8 23.3 27.7 Switzerland BVG 2015 22.4 25.6 24.4 27.7 United Kingdom S2PMA/S2PFA (CMI 2015 1.25%) 22.2 25.9 24.3 28.1 United States RP-2006* (MP2016) 20.9 24.2 22.9 26.1

* RP-2006 refers to the RP-2014 base rates projected backward to the year 2006 with Scale MP-2014

For More Information If you would like further information on accounting assumptions for defined benefit plans, contact your local Aon Hewitt actuarial consultant, Kirsten Miller in Glasgow, Scotland (+44.7834.609704), or Yc Tao in New York, New York (+1.212.441.2556).

For more information on the countries in this newsletter, please refer to the Aon Hewitt Country Profiles. You can learn more about the Country Profiles here.

If you are a subscriber to the Aon Hewitt Country Profiles platform and wish to access the full text of any Country Profile including all Updates, please click here and enter your Country Profiles User Name and Password.

About Aon Aon plc (NYSE:AON) is a leading global professional services firm providing a broad range of risk, retirement and health solutions. Our 50,000 colleagues in 120 countries empower results for clients by using proprietary data and analytics to deliver insights that reduce volatility and improve performance.

For further information on our capabilities and to learn how we empower results for clients, please visit http://aon.mediaroom.com.

© 2017 Aon plc This document is intended for general information purposes only and should not be construed as advice or opinions on any specific facts or circumstances. The comments in this summary are based upon Aon Hewitt’s preliminary analysis of publicly available information. The content of this document is made available on an “as is” basis, without warranty of any kind. Aon Hewitt disclaims any legal liability to any person or organization for loss or damage caused by or resulting from any reliance placed on that content. Aon Hewitt reserves all rights to the content of this document.

http://www.aon.com/attachments/human-capital-consulting/Aon_Hewitt_Country_Profiles_Brochure_2015.pdf

https://was2.hewitt.com/loginpanels/TOC-pvm/pvm6login.fcc?TYPE=33554433&REALMOID=06-6b9ff8e4-0a27-1014-ad12-83c9c005fd42&GUID=0&SMAUTHREASON=5&METHOD=GET&SMAGENTNAME=-SM-FkS7IOiQHTNREZ4BoQABYN9HB4adVUwoJua9%2bw00zzY2PXLmkRzD6JtkE2NaRZNk&TARGET=-SM-HTTPS%3a%2f%2fwas2%2ehewitt%2ecom%2fpvm%2fcommon%2fco_login%2ejsp%3fsitecode%3dcp

http://aon.mediaroom.com/

Documents

Global Report: Global Survey of Retirement Plan … · Global Report: Global Survey of Retirement Plan Accounting Assumptions . ... Canadian AA bond universeand created challenges