107
Notes to the Consolidated Financial Statements (continued)
Inflation risk
Life expectancy
Principal actuarial assumptions as at the balance sheet date
-
-
Sensitivity analysis
Heineken NV.
Report of the
Report of the
Financial
Sustainability
Other
Annual Report 2016
Introduction
Executive Board
Supervisory Board
Statements
Review
Information
Some of the pension obligations are linked to inflation. Higher inflation will lead to higher liabilities, although in most cases caps on the level of
inflationary increases are in place to protect the plan against extreme inflation. The majority of the plan assets are either unaffected by or loosely
correlated with inflation, meaning that an increase in inflation will increase the deficit.
HEINEKEN provides employees in the Netherlands with an average pay pension plan, whereby indexation of accrued benefits is conditional on the
funded status of the pension fund. In the UK, inflation is partly managed through the use of a mixture of inflation-linked derivative instruments.
These instruments match 41% of the inflation-linked liabilities (2015: 39%).
The majority of the plans' obligations are to provide benefits for the life of the member, so increases in life expectancy will result in an increase in the plans'
liabilities. This is particularly significant in the UK plan, where inflation-linked increases result in higher sensitivity to changes in life expectancy. In 2015,
the Trustee of SNPP implemented a longevity hedge to remove the risk of a higher increase in life expectancy than anticipated for current pensioners.
Based on the significance of the Dutch and UK pension plans compared with the other plans, the table below only includes the major actuarial
assumptions for those two plans as at 31 December:
The Netherlands UK*
In
2016
2015
2016
2015
Discount rate as at 31 December
1.5
2.3
2.7
3.9
Future salary increases
2.0
2.0
Future pension increases
0.4
0.9
3.1
3.0
The UK plan closed for future accrual, leading to certain
assumptions being equal to zero.
For the other defined benefit plans, the following actuarial assumptions apply at 31 December:
Europe Americas
Africa, Middle East Eastern Europe
In
2016
2015
2016
2015
2016
2015
Discount rate as at 31 December
0.6-6.8
0.8-2.3
7.0-7.6
7.0
1.5-15.5
12.0
Future salary increases
0.0-3.5
0.0-3.5
0.0-4.5
4.5
0.0-5.0
7.5
Future pension increases
0.0-1.5
0.0-1.2
0.0-3.5
3.5
0.0-3.5
3.0
Medical cost trend rate
0.0-4.5
0.0-4.5
0.0-5.0
5.1
0.0-5.0
4.5
Assumptions regarding future mortality rates are based on published statistics and mortality tables. For the Netherlands, the rates are obtained
from the AG-Prognosetafel 2016', fully generational. Correction factors from Towers Watson are applied on these rates. For the UK, the future
mortality rates are obtained by applying the Continuous Mortality Investigation 2014 projection model with an assumed long term rate of 1.5% p.a.
to the Self-Administered Pension Schemes Series 2 (year of birth) tables with a 112% (male)/109% (female) weighting for pensioners and a 105%
(male)/106% (female) weighting for non-pensioners.
The weighted average duration of the defined benefit obligation at the end of the reporting period is 18 years.
HEINEKEN expects the 2017 contributions to be paid for the defined benefit plans to be in line with 2016.
Reasonably possible changes at the reporting date to one of the relevant actuarial assumptions, holding other assumptions constant, would have
affected the defined benefit obligation by the amounts shown below:
31 December 2016
31 December 2015
Effect in millions of EUR
Increase in
assumption
Decrease in
assumption
Increase in
assumption
Decrease in
assumption
Discount rate (0.5% movement)
(695)
798
(677)
771
Future salary growth (0.25% movement)
23
(22)
21
(20)
Future pension growth (0.25% movement)
332
(309)
300
(292)
Medical cost trend rate (0.5% movement)
5
(4)
6
(5)
Life expectancy (1 year)
300
(301)
287
(290)
Although the analysis does not take account of the full distribution of cash flows expected under the plan, it does provide an approximation of the
sensitivity of the assumptions shown.