Explanatory Memorandum to COM(2008)443 - Annex XI to the Staff Regulations - Main contents
Please note
This page contains a limited version of this dossier in the EU Monitor.
| dossier | COM(2008)443 - Annex XI to the Staff Regulations. |
|---|---|
| source | COM(2008)443  |
| date | 10-07-2008 |
|
52008DC0443
Report from the Commission to the European Parliament AND to the Council on Annex XI to the Staff Regulations /* COM/2008/0443 final */
[afbeelding - zie origineel document] COMMISSION OF THE EUROPEAN COMMUNITIES
COM(2008) 443 final
REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT AND TO THE COUNCIL
on Annex XI to the Staff Regulations
REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT AND TO THE COUNCIL
on Annex XI to the Staff Regulations
Article 15 i of Annex XI to the Staff Regulations provides that the provisions of Annex XI are to be reviewed at the end of the fourth year after 1 July 2004, particularly in the light of their budgetary implications. To this end, the Commission must submit a report to the European Parliament and the Council and, where appropriate, a proposal to amend this Annex on the basis of Article 283 of the EC Treaty.
This report implements the aforementioned provision. It focuses on the issues, whether Annex XI serves its purpose and whether any amendments are needed to ensure smooth application of the 'method'.
The provisions of the current method for adjusting remuneration and pensions apply from 1 July 2004 until 31 December 2012 and are expressed in Articles 64, 65 and 65a of the Staff Regulations and Annex XI.
The main objectives of the method are:
- automatic salary adjustment i in order to avoid that the work of all Community Institutions and Agencies is disrupted by annual negotiations and social unrest;
- transparent, efficient, relatively straightforward rules to determine salary adjustments for officials and other servants of all EU Institutions..
In order to ensure proper functioning of the method, the following principles have been laid down:
- parallelism with national officials in the evolution of purchasing power;
- equality of purchasing power among civil servants in different duty stations.
This report analyses whether the method laid down in Annex XI to the Staff Regulations follows the principles and reaches the objectives mentioned above.
The Specific Indicator measures changes in the purchasing power of civil servants in central governments of Member States.
A full time series of data for each of the 27 EU Member States is not available (Bulgaria and Romania joined with effect from 2007, and Estonia has not supplied data). The TABLE 1 summarises the available data for each country for the period 2004-2007, together with a simple arithmetic mean value for the period. The next table shows the same data, re-expressed as a cumulative index (2003=100).
Under the May 2004 Staff Regulations, the global specific indicator is computed using a sample of eight reference countries, rather than using data for all the Member States of the EU. The eight countries are: Belgium, Germany, France, Spain, Italy, Luxembourg, Netherlands and the United Kingdom. This choice was based on the estimate that salary development in the sample countries would be similar to the one in EU15.
Prior to 2004, the global specific indicator was calculated by reference to the EU as a whole (i.e. EU12 1987-1994, EU15 1995-2003).
The 2007 enlargement and recent economic development reduced the EU GDP percentage of this sample over time and it is likely that in the coming years this trend will continue. Annex XI lays down a simplified procedure for the Council acting on a Commission proposal under Article 65 i of the Staff Regulations to adopt a new sample. Such a sample should represent at least 75% of the European Union GDP.
TABLE 2 shows the time series for these eight countries since 2004, together with a simple arithmetic mean value for the period. TABLE 3 shows the same data, re-expressed as a cumulative index (2003=100).
It is apparent that national civil servants' net salaries in real terms have decreased over this period by around 6% in Germany, by around 3% in Luxembourg and by around 1% in Spain. Over the same period, they have increased by around 1% in Italy and by around 2% in the Netherlands and the United Kingdom.
GRAPH 4 shows the time series since 2004 of the global specific indicator computed using data for these eight reference countries (weighted by GDP), together with the Brussels International Index and the consequent annual adjustment for EU staff.
TABLE 5 presents these time series data, re-expressed as a cumulative index over the period. The global specific indicator, computed as a simple average for the four-year period, is -0.3%; the cumulative impact to July 2007 (2003=100) is -1.0%. The average annual inflation in Brussels over this period was +1.9%; the cumulative impact is +7.8%. The average annual adjustment to remuneration and pensions was +1.7%; the cumulative impact is +6.8%. The cumulative impact is illustrated in GRAPH 6. The cumulative negative specific indicator (-1%) resulted in a decrease of purchasing power of EU civil servants.
A range of alternative calculations are possible for the global specific indicator. These test calculations carried out by Eurostat include:
- A simple average of Brussels and Luxembourg values (on the logic that this reflects the location of the headquarters of the EU institutions)
- Calculations for EU25
- EU15, euro area and other country groupings
- Alternative GDP weightings
- Weighting country values using national population totals
- Weighting country values by population of national civil servants in central public administration
- Weighting country values using Qualified Majority Voting rights.
A hypothetical global specific indicator calculated using available data for all member states (i.e. EU25 in absence of time series data for Bulgaria and Romania, and excluding Estonia in absence of data) would be quite different from the one calculated for the sample of 8 countries. The simple average for the 4 year period would be +0.2% instead of -0.3%; the cumulative impact to July 2007 (2003=100) would be +0.8% instead of -1.0%.
The current methodology aims to monitor the evolution of net salaries of national civil servants in an attempt to ensure equivalence of the purchasing power of net EU staff salaries. It is important to note that the net salaries of national civil servants already take into account the impact of pension contributions, medical insurance, etc. payable in the Member States. Where there is a change in such compulsory deductions for one or more countries, this would be reflected in the measured evolution of the net salary, which is a key indicator for the annual increase in basic pay of EU officials. However, where there is a change in the coverage of risks, it is not taken into account.
(a) Social contributions paid by EU officials
EU officials pay obligatory social contributions, which are deducted from their take-home salary. These include pension contributions and medical (health and accident) insurance contributions. The contributions are periodically increased, and are set entirely independently of the reference countries' contributions. There is consequently an element of double-counting in the current system.
TABLE 7 attempts to quantify the cumulative impact of departure from parallelism due to social contributions (situation at May 2004 = 100), by deducting the impact of changes in the pension contribution rate payable by staff from the annual adjustment, to identify the actual combined effect.
Over the period 2004-2007, the actual cumulative salary adjustment is around 0.7 per cent lower than the published annual adjustment, due to the further impact of changes in the pension contribution rate.
(b) Risks coverage
National civil servants may pay various social contributions which guarantee them protection from certain risks. The risk coverage in each Member State is specific to that country. Similarly, only a specific set of risks are covered by the EU welfare system. These differences in coverage and the impact of any changes in coverage make comparison difficult, and no attempt is made to correct this. Nevertheless, this situation also creates an element of departure from parallelism. The salaries of EU staff are affected by changes in social contributions for risk coverage in Member States, protection to which they themselves may not be entitled. In particular, they are penalised if national civil servants' contributions are increased, and they benefit if national civil servants' contributions are decreased.
The special levy is an additional deduction from the remuneration of EU officials. It introduces another element of double counting.
TABLE 8 attempts to quantify the cumulative impact of departure from parallelism due to the special levy (situation at May 2004 = 100), by deducting the impact of changes in the special levy from the annual adjustment, to identify the actual combined effect.
Over the period 2004-2007, the actual cumulative salary adjustment is 0.6 per cent lower than the published annual adjustment, due to the further impact of the introduction and the (ongoing) annual changes in the special levy payable by staff.
TABLE 9 indicates the combined additional impact on the annual adjustment of the evolution of pension contributions and the evolution of the special levy since 2004.
The cumulative impact is shown in GRAPH 10. If the cumulative change in salaries since July 2003 of +6.8% (i.e. annual adjustments 2004-2007) is adapted to take into account the cumulative impact of the changes in the pension contribution rate and the changes in the special levy since May 2004, the actual overall cumulative change is only +5.4%, i.e. 1.3 per cent lower. This is less than the cumulative inflation in Brussels for the same period, of +7.8%.
Member States may experience some practical difficulties in supplying Eurostat with the necessary information. Measurement difficulties could include applying the correct definition of gross pay; relating pay information to the correct reference period; collecting information for the correct reference population within government; transmitting this information to Eurostat. These issues are discussed in more detail below.
The current methodological guidance states that all elements of remuneration which affect the purchasing power of national civil servants should be taken into account when calculating the gross remuneration for transmission to Eurostat. In addition to basic salaries, all general bonuses and premiums should be reported, including non-pensionable lump-sum amounts, together with allowances (e.g. child benefit, family allowances). For measurement reasons, non-monetary income components are typically excluded. Occasional interpretation issues arising in the past have been clarified bilaterally on a case-by-case basis. However, some standardised methodological guidance is now provided i on the interpretation of components of gross pay, treatment of irregular payments, identification of reference population in order to assist data suppliers with a more detailed understanding of the source information to be calculated and supplied to Eurostat, and permit end-users to have a more detailed understanding of indicators produced by Eurostat.
Issue of statutory contributions becoming voluntary
The issue of dealing with the transition of some contributions from statutory to voluntary arose in 2006 when one Member State communicated data with two sets of figures for the previous year. The need of two sets of figures for previous year, as explained by the Member State, was due to the changes in the health insurance system when some statutory contributions became voluntary giving an option for civil servants to choose from three standard packages. The Member State claimed that the average package should be considered as statutory; however, the Commission considered that the least costly package which allows them to comply with the legislation in force shall be considered as statutory, as civil servants can choose other packages on voluntary basis. This interpretation gained the support of the Council Legal Service, which declared that the Commission proposal complied with the relevant provisions of the Staff Regulations and the request of the Member State gave rise to a number of specific reservations which it set out in the opinion.
As there were diverging opinions on this issue in the Council Working Group on the Staff Regulations, the Working Group asked the Commission to examine it in the framework of the Article 65 Working Group in order to reach a joint understanding. In the Article 65 Working Group, the Commission proposed the text for the joint understanding, which was endorsed by the majority of the members in this Working Group. The text was subsequently integrated into the method manual i adopted by the Art 65 WG in March 2008:
- A change in the social security system or taxation regime is not considered to be a change in the salary scale. Changes between year t-1 and year t are taken into account in the calculation of the indicator for the current year.
- In the case where statutory deductions increase or new ones are introduced, this will be reflected in a decrease of net salaries used to compute the specific indicator. Similarly, where statutory deductions decrease or are withdrawn, this will be reflected in an increase of net salaries.
- In the case where a voluntary social contribution becomes a statutory deduction, this will be taken into account by decreasing net salaries. Where a statutory deduction becomes voluntary, this will no longer be taken into account.
Where a choice amongst different packages is possible, the least costly package allowing minimum compliance with legislation in force shall be considered as statutory.
In itself a change of salary scale does not necessarily impact on the pay of current staff. Some grades disappear, others are introduced, existing staff are reallocated to new grades. A change in the social security system or taxation regime is not considered to be a change in salary scale.
Such large-scale changes are rare events. They do, however, raise a practical problem: how to compare the information available at time (t) and identify the evolution from time (t-1).
A practical solution to this problem was proposed to the Article 65 Working Group and was subsequently integrated in the method manual i.
Other things being equal, the key principle is that immediately prior to the change and immediately after the change, the total number of employees and the total salary bill are unchanged.
2.5.4. Samples covering less than 100%
Selected grades for each functional category (administrators; assistants) are required to be representative of the category (e.g. modal) and to cover at least 75% of actual employment numbers. In practice, most countries supply data covering 100% of staff.
A problem potentially arises where it is proposed to change the sample composition, as this could impact on the weighted average salary used at time (t) and the comparison with (t-1).
A practical solution was proposed to the Article 65 Working Group and was subsequently integrated in the method manual i.
With a given distribution of staff numbers and salaries, different samples could initially be selected, both of which respect the requirement to cover at least 75% of employees. If, twelve months later, the sample of grades is changed (e.g. following substantial movements of staff), this leads to the logical conclusion that, where the decision is taken not to cover 100% of employees and having once selected a representative sample, that sample should then be retained without modification for future comparisons. Any proposal to change the sample should be signalled in advance and only agreed in consultation with Eurostat.
At some stage during or after the process it is conceivable that an error may be identified either by a Member State or by Eurostat. Such errors must be officially communicated (Eurostat to Member State or the other way round). This then raises budgetary and administrative questions about whether/how to adjust such corrections retrospectively.
Where errors are detected, the Staff Regulations impose no limit on retrospective adjustment. However, to limit the practical administrative and financial burdens, a gentleman's agreement was reached in January 2005 and is now integrated in the method manual i. This policy provides for revision in the following circumstances:
- If identified prior to Council adoption, the Eurostat Report will be revised and a new draft Regulation will be submitted to the Council.
- For a period of six months after the adoption of this Regulation there is a possibility to start a revision of the adjustment (which can result in a retroactive correction).
- After this six-month period, data for year t are considered as final.
A negative legal opinion was presented to the Council in November 2006, concerning the existence of more than one dataset for any given year. If a dataset for a given year is found to be incorrect, provision should be made for a formal revision.
In its opinion i the Council Legal Service made several important clarifications:
- When Eurostat calculates the specific indicators for these purposes, the only data to be taken into account for the previous year are those used for the adjustment exercise for the year in question.
- It would be possible, for duly justified reasons, to alter the statistical data used for a given year, but this would make it necessary to carry out a correction, following the applicable procedure, of the results of the calculation for which these data were used – i.e. amending the Regulation concerning the annual adjustment of the remuneration of officials for the previous year.
- In the case of statutory deductions become becoming non-compulsory, the Council cannot take account of these data as if the deductions were compulsory.
The Brussels International Index (BII) is simply a measure of consumer price inflation in Brussels. It is a Laspeyres price index, computed as a weighted average of the component price indices, using Family Budget Survey expenditure patterns of EU civil servants in Brussels as the weights.
CHART 11 shows the BII time series (annual increase June-June by reference to previous year), together with comparable information for Belgian HICP.
Differences in the trend are due to the use of different index sources for certain basic headings, and the different weights used for aggregation.
The average BII over the period from 2004 to 2007 has been 101.9. The cumulative index for the whole period to June 2007 (base June 2003 = 100) is 107.8.
By comparison, the average HICP in Belgium over the period was 102.2 and the cumulative total for the period was 109.1.
The remuneration of officials serving in places of employment outside Brussels and Luxembourg is expressed in euro and adjusted by a correction coefficient for their location which is set above, below or equal to 100%.
Correction coefficients are mathematical factors which, when applied to a monetary amount expressed in euro (e.g. Brussels salary), together with the official exchange rate to the euro (e.g. for a particular duty station city), identify the economic parity, which is a statistical value reflecting the cost-of-living difference (e.g. between the duty station city and Brussels). The economic parity is the average ratio of prices. It thus reflects the amount in national currency in a particular location which is needed to purchase the equivalent basket of goods and services in Brussels with one euro.
TABLE 12 shows correction coefficient values for individual duty stations for the period 2003-2007.
From this table it is apparent that the correction coefficients for different locations have followed different trends over time. In 10 locations they have decreased over the period, whilst 20 locations have experienced increases.
Some care is needed when interpreting these data, because the underlying economic parities are designed to give a robust estimate of spatial differences at each point in time, rather than a time series evolution. Some observations can nevertheless be made.
The places of employment with the largest number of staff are Brussels and Luxembourg. The next duty station locations in rank order by numbers of staff affected by correction coefficients would be: Varese (IT) (2.5%), Alicante (ES) (1.6%), Karlsruhe/Frankfurt (DE) (1.2%), Dublin (IE) (0.7%) and The Hague (NL) (0.6%).
Over time, comparison of consumer expenditure patterns for the national population and the spending pattern of EU officials has repeatedly demonstrated that expatriate staff can differ markedly in their purchasing habits. In consequence it is not appropriate to use national figures to aggregate the economic parities to produce an overall value. Instead, information has to be collected from EU officials in the various duty stations.
These consumer expenditure weightings are obtained from Family Budget Surveys (FBS) conducted at periodic intervals (5 to 7 years). Whilst the target populations and sample sizes are clearly smaller, the percentage response rate is comparable with Household Budget Surveys conducted at national level.
Price data from the European Comparison Programme are considered to be robust. However, the review process to establish data which are suitable for computing multilateral purchasing power parities logically focuses on that outcome, and not on data suitability to establish bilateral purchasing power parities with Brussels. Some re-examination of underlying data is therefore desirable, and could improve the quality of bilateral parities.
From TABLE 12 it is clear that for certain duty station locations, notably those in certain new Member States, the apparent evolution in price levels relative to Brussels has been particularly significant. Standard reasons for this are described and quantified in the Eurostat Annual Report.
GRAPH 13 shows the overall PPP evolution for the new Member States for the period 2004-2007 (2003=100). From this graph it is clear that the global parity has decreased for the period 2003-2007 by 15-25% in Czech Republic and Slovakia, and by 5-10% in Lithuania and Poland, with a decrease overall of less than 5% in Malta. By contrast the parity movement has been positive in Estonia, Cyprus, Hungary, and has exceeded 5% in Slovenia and Latvia.
The presentation of correction coefficient results for July 2007 has generated particularly high numbers of queries relating to duty stations in certain of the new Member States, in Varese, Italy and in Spain. Staff representatives in Luxembourg have also raised queries about the cost-of-living difference relative to Brussels.
Particular efforts have been made during 2007 to clarify and explain methodology, both to staff representatives (during the social dialogue process) and in the form of extensive correspondence with delegations, for whom a standardised presentation has been developed.
To help decipher trends in data series, analysts often apply various “smoothing techniques” designed to reduce or eliminate volatility in data.
The “moving average” is one simple smoothing technique, which works by consolidating the periodic data points into longer units of time – for example an average of several years’ data. The larger the number of years included in the average, the smoother the series.
There is a downside to using a moving average to smooth a data series, however: because the calculation relies on historical data, some of the variable’s timeliness is lost. And the application of a moving average could cause discrepancies in the equality of purchasing power in different duty stations.
Other techniques are being explored in the Article 64 Working Group.
Contents
- budgetary implications
- Brussels, 10.7.2008
- 1. INTRODUCTION
- 1.1. Basic objectives and principles of the method
- 2. SPECIFIC INDICATOR
- 2.1. Specific indicator trends for individual countries
- 2.2. Trend of the global specific indicator as currently defined
- Annual and cumulative data (by country)
- Global specific indicator for eight countries
- 2.3. Other possible approaches to the computation of global specific indicators
- 2.4. Departure from the general principle of parallelism
- 2.4.1. The impact of EU social contributions
- 2.4.2. The impact of the EU special levy
- 2.4.3. Combined impact of EU social contributions and the EU special levy
- 2.5. Problems encountered and solutions implemented or proposed
- 2.5.1. Availability of data and metadata
- Measurement difficulties with certain components of gross pay and their definition
- 2.5.2. Interpretation of the Staff Regulations
- 2.5.3. Changes to salary scale
- 2.5.5. Retrospective revision of figures
- 2.5.6. Several datasets for the same year
- 3. THE BRUSSELS INTERNATIONAL INDEX
- 4. CORRECTION COEFFICIENTS
- 4.1. Trends of correction coefficients
- 4.2. Problems encountered and solutions implemented or proposed
- 4.2.1. Obtaining robust consumer expenditure weightings (small populations)
- 4.2.2. Obtaining robust survey parities
- 4.2.3. Enlargement (transition from extra-EU to intra-EU)
- 4.2.4. Misunderstandings by end-users
- 4.3. Smoothing techniques (e.g. moving average)
- OF THE FUNCTIONING OF THE METHOD FOR ADJUSTMENT OF REMUNERATION AND PENSIONS
- The results of the method
- Difference EU Officials / EU8 national officials
- Net Pay increases including changes in special levy and the pension contribution rates
- Net pay difference EU Officials / EU8 national officials
- Amounts in EUR million at current prices
- Heading 5 expected margin
- 6. CONCLUSIONS
- ANNEX
- TABLE 3
- GRAPH 4
- TABLE 5
- GRAPH 6
- TABLE 7
- TABLE 8
- TABLE 9
- GRAPH 10
- CHART 11
- TABLE 12
- GRAPH 13
The method has fulfilled its aims because the EU officials pay has followed the national official pay increases in real terms.
EU Officials Basic Pay (and pensions)
Increase at current prices 0,7% 2,2% 2,3% 1,4%
Brussels International Index 1,9% 2,2% 2,1% 1,4%
Basic pay increase in real terms -1,2% 0,0% 0,2% 0,0%
Cumulated increase in real terms -1,2% -1,2% -1,0% -1,0%
Pay increase in real terms
EU Officials basic Pay -1,2% 0,0% 0,2% 0,0%
National Officials net pay (EU8 average) -1,2% 0,0% 0,2% 0,0%
Difference EU Officials / EU8 national officials 0,0% 0,0% 0,0% 0,0%
Cumulated Difference EU Officials / EU 0,0% 0,0% 0,0% 0,0%
Including the changes in the pension contribution and in the special levy rates, EU officials have had a lower pay increase in real terms than national officials.
EU Officials Net Pay
Basic Pay increase in real terms -1,20% 0,00% 0,20% 0,00%
Net effect of changes in the special levy rate -0,2% -0,2% -0,2%
Net effect of pension contribution rate changes -0,35% -0,35%
Net Pay increase in real terms -1,55% -0,55% 0,0% -0,2%
Cumulated net pay increase in real terms -1,5% -2,1% -2,1% -2,3%
Net Pay increase in real terms
EU Officials net Pay -1,55% -0,55% 0,0% -0,2%
National Officials net pay (EU8 average) -1,2% 0,0% 0,2% 0,0%
Difference EU Officials / EU8 national officials -0,35% -0,55% -0,2% -0,2%
Cumulated Difference EU Officials / EU -0,3% -0,9% -1,1% -1,3%
The expected evolution of administrative expenditure, which is dominated by pay and pension's expenditure, is compatible with the financial framework.
The Heading 5 financial programming table below(extracted from 2009 Preliminary draft Budget doc V):
Commission without pensions 3 3 3 3 3 4 4
Pensions 1 1 1 1 1 1
Other institutions excluding pensions 2 2 2 2 3 3 3
Total heading 5 * 6 7 7 8 8 8 9
Heading 5 ceiling with footnote** 7 7 7 8 8 8 9
* The difference between the total and the sum are due to rounding. ** The H5 ceiling includes the staff contribution to the retirement scheme
The system of remuneration laid down in the Staff Regulations guarantees equivalence of purchasing power between Community officials, taking account of their family situation and the cost of living in their place of employment.
The 2004 method was the result of negotiations between the Council, the administrations of the Community institutions and representatives of the staff of the institutions. It is applicable from 1 July 2004 to 31 December 2012, and has produced the following results:
1. Since 2004, national civil servants of eight reference Member States have seen the purchasing power of their salaries fall (the cumulative global specific indicator for 2004-2007 is -1.0%).
2. The use of data for alternative country groupings, or using alternative weights or using alternative definition of salaries would produce different figures. For example, the equivalent cumulative global specific indicator for 2004-2007 for the EU25 would be +0.8%.
3. The principle of parallelism inherent in the method has automatically applied this reduction (-1.0%) to the net salaries of Community officials.
4. The increase of the pension contribution rate and the changes in the special levy resulted in an additional loss of purchasing power (-1.3%).
5. The combined effect has thus been a loss of purchasing power for Community officials (-2.3% in four years).
6. The budgetary impact of Annex XI for the period 2004-2007 was a reduction of 1% in real terms on pay and pensions expenditure.
7. Only the fact that the method works automatically has enabled conflicts over annual negotiations to be avoided.
8. It can be said that Community officials' salaries have moved in line with changes in the economic and social situation, with a loss of purchasing power going beyond parallelism with national civil servants.
9. Several problematic issues were identified by Eurostat. However, they were addressed or will be addressed in the Article 64 and Article 65 Working Groups and do not therefore require modification of the Staff Regulations.
10. The percentage of total EU GDP represented by the present reference sample for the calculation of the specific indicator has decreased over the period 2004-2007 and is approaching 75%. There is not an immediate need to update this sample. Nevertheless the Commission may consider proposing an update in the coming years.
For the above reasons, the Commission takes the view that the method has achieved its objectives and followed the principles laid down in Annex XI and that improvements are possible through actions that do not imply an immediate revision of Annex XI. The Commission therefore considers that no modification of the Staff Regulations is needed.
TABLE 1
[afbeelding - zie origineel document] [afbeelding - zie origineel document]
TABLE 2 [afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]
[afbeelding - zie origineel document]