[pic] | COMMISSION OF THE EUROPEAN COMMUNITIES |

Brussels, 21.10.2005

SEC (2005)1329

COMMISSION STAFF WORKING DOCUMENT

Annex to the Communication on an EU common methodology for assessing administrative costs imposed by legislation Outline of the proposed EU common methodology and Report on the Pilot Phase (April– September 2005) {COM(2005)518 final}

COMMISSION STAFF WORKING DOCUMENT

Outline of the proposed EU common methodology and Report on the Pilot Phase (April– September 2005)

Outline of EU common methodology for assessing administrative costs imposed by legislation 4

1. Aims and purposes 4

2. Definition of administrative costs 4

3. Core equation of the cost model 5

4. Scope and frequency of estimates 5

5. Expected level of accuracy 5

6. Data sources 6

7. Determination of the regulatory origin 6

8. Reporting 6

9. Methodological caveats 7

Report on the Pilot Phase (April – September 2005) 9

1. Background 9

2. Constraints 9

3. Methodological findings 10

3.1 Aims and purposes of the model 10

3.2 Definition of administrative costs 11

3.3 Core equation of the cost model 12

3.4 Scope and frequency of estimates 13

3.5 Expected level of accuracy 14

3.6 Data sources 14

3.7 Division of responsibility 17

3.8 Reporting 18

3.9 Methodological caveats 19

4. Feasibility of an EU common methodology 19

4.1 List of methodological and technical issues to be further examined 20

4.2 Costs of detailed quantification of administrative burden 21

4.2.1 Workload for Commission services 21

4.2.2 Workload for national authorities 22

4.2.3 Workload for private entities 23

4.3 Contribution of Member States 23

5. Added value / Does analytical quantification help meeting better regulation principles? 23

5.1 Analytical quantification helps in assessing measures from the point of view of those affected 23

5.2 Analytical quantification contributes to transparency 24

5.3 Analytical quantification provides an interesting indicator for monitoring progress and could help with the reduction of costs 24

6. Conclusions 24

Outline of EU common methodology for assessing administrative costs imposed by legislation

Building on the experience of the Member States and its own practice, the Commission outlined in March 2005 a model encompassing methodological elements enjoying wide consensus and backed by solid empirical evidence, but also elements whose feasibility or value-added for a common EU approach had yet to be demonstrated[1]. The Commission services have now tested many of those elements.

Having concluded on the findings of the pilot conducted between April and September 2005 (see Report in Part II of this Staff Working Document), Commission services are of the opinion that some features envisaged in the ‘EU Net Administrative Costs Model’ of March 2005 should be kept while others appear as unduly complex or subjective and should be discarded. The Commission services also believe that pending methodological issues are either minor or a matter of optimisation, and therefore that they do not constitute an obstacle to the introduction of an EU common methodology.

Taking into account the findings of the pilot phase and the remaining uncertainties, a future EU common approach for assessing administrative burden could be based on the following principles and features[2].

Aims and purposes

The main aim of the EU common methodology is to assess the net costs of administrative obligations imposed on enterprises, the voluntary sector, public authorities and citizens (net costs = costs introduced by legislation minus the costs eliminated by legislation at EU and/or national level). It will also allow the origin of administrative obligations to be determined, distinguishing between regional, national, EU and international obligations.

The methodology is intended for microeconomic purposes, i.e. to assess the administrative costs imposed by individual pieces of legislation. It should be used in the impact assessment of a proposed measure (ex ante) and for the simplification of an existing measure (ex post).

Definition of administrative costs

Administrative costs are defined as the costs incurred by enterprises, the voluntary sector, public authorities and citizens in meeting legal obligations to provide information on their action or production, either to public authorities or to private parties. Information is to be construed in a broad sense, i.e. including costs of labelling, reporting, monitoring and assessment needed to provide the information and registration.

Recurring costs and, where significant, one-off costs have to be taken into account.

In order to keep the EU common methodology as simple as possible and to minimise subjective judgment in the assessment, no distinction should be made between ‘pure obligation’ and obligation corresponding to what an entity would normally do in the absence of any legal obligation. Similarly, optional participation which could be considered as de facto obligatory should not be assimilated to administrative burden. Member States wishing to assess the costs of optional schemes may do so but should present these costs as a separate item.

Core equation of the cost model

Administrative costs should be assessed on the basis of the average cost of the required action (Price) multiplied by the total number of actions performed per year (Quantity). The average cost per action will be generally estimated by multiplying a tariff (based on average labour cost per hour including prorated overheads) and the time required per action. Where appropriate, other types of costs such as equipment costs will be taken into account. The quantity will always be based on the frequency of required actions multiplied by the number of entities concerned.

Σ P x Q (Price = Tariff x Time; Q: Quantity = Number of businesses x Frequency).

Scope and frequency of estimates

The EU common methodology must be applied in a proportionate manner. It should only be applied when the scale of the administrative obligations imposed by an EU act justifies it and the effort of assessment should remain proportionate to the scale of the administrative costs imposed by the legislation.

Indicative thresholds should ideally be set to determine de minimis costs that do not need to be assessed. For administrative obligations requiring little equipment, if the amount of time per action is small and the frequency low, the obligation does not need to be quantified. What is to be considered as negligible or very low frequency needs to be further tested.

The frequency with which the administrative costs imposed by a specific piece of legislation are reviewed will be defined on a case-by-case basis.

Expected level of accuracy

In order to keep assessment of costs at a reasonable level and ensure compatibility with national methodologies, the EU common methodology will be based on working assumptions simplifying the complex reality of the Union. Among other things, the methodology will normally assume that interested entities fully comply with EU legislation. Assessment will also be based on ideal types (typical firms, typical public service, etc.). Consequently, the proposed instrument will only be able to approximate rather than measure.

Ambition as to the level of accuracy of these approximations will depend on the level of administrative costs induced by the (proposed) measure, the cost of the assessment and its potential benefits (proportionate analysis) and, last but not least, the input provided by Member States.

The expected level of accuracy will also depend on the type of act measured. It will normally be lower in the case of ex ante assessment of EU directives. Such assessment is complicated by the fact that EU directives have to be transposed into Member States’ national legislation and that Member States rarely decide on how to transpose until Community law is adopted. The degree of cooperation of Member States will therefore be crucial for the assessment. In default of detailed contributions from Member States, the assessment will have to be based on hypothetical transposition measures. The latter option is likely to increase the margin of error. Such estimates will therefore have to be interpreted with care.

In order to avoid spurious accuracy, impact assessments and ex post evaluations may continue to refer to range of estimates when presenting their findings, but should only use the median figure when summarising findings.

An average margin of error should be progressively determined by comparing ex ante and ex post quantifications of the same act based on the EU common methodology. Other ways to determine margins of error should be investigated in parallel.

DATA SOURCES

Data collection methods that might be chosen according to the individual case include: focus groups, wide consultation of stakeholders, field trials, consultancy studies, extrapolation from comparable situations and expert assessment. The choice of method lies with the institution or Member State carrying out the assessment.

In standard cases, it will be sufficient to produce rough estimates based on: available EU statistics (provided, among others, by Eurostat and the Small and Medium-Sized Enterprises Observatory); standard ratios (for example assessing overheads on the basis of a mark-up percentage on labour costs or discounting costs of legal obligations corresponding to “normal business operation”); the opinion of experts; and Member State studies.

In exceptional cases, field work limited to a sample of Member States and/or questionnaires sent to a standard sample of the business community and simulation may have to be used.

DETERMINATION OF THE REGULATORY ORIGIN

The EU common methodology will distinguish between international, EU, national and regional (i.e. subnational) administrative obligations. The origin of administrative burdens should be determined on the basis of a common set of questions.

REPORTING

Estimates of administrative costs need to be reported in a standardised manner to allow for their comparison and addition. The same report sheet must therefore be used by all when assessing or contributing to the assessment of administrative costs imposed by EU legislation (see below). Users may however add ad hoc information to the report sheet, as long as this does not alter the standard part of the report sheet[3].

For proposals such as the thematic strategies, the common report sheet will often act as a summary of more detailed analyses. The existence of a report sheet will not prevent services from presenting more detailed data in separate tables and texts.

For a regulatory act transposing an international or EU act, the report sheet must include the name and reference of that international or EU act, as well as a simple concordance table made of two columns (the first column gives the reference of the article detailing the obligation assessed; the second column gives the reference of the ‘original’ obligation, i.e. the article of the act laying down the obligation transposed by the act being assessed).

The sheet will report on the nature of the administrative obligation based on a basic common typology. In order to report on who will bear the costs, answers for the ‘target group’ column will also be based on a basic typology. When a measure amends existing provisions and if it removes administrative obligations, the sheet will include negative figures corresponding to the burden reduction. Finally reporting on regulatory origin should be done in a standardised way, referring to four different levels: international, EU, national and regional (i.e. subnational) obligations.

Methodological caveats

When reporting on their assessment, EU institutions and Member States will indicate, succinctly but clearly, their working assumptions and methodological limitations. This will include assumptions concerning compliance rate, warning about the nature of the data presented (estimates and not exact measures); and indication of the margin of error.

Reporting sheet proposed for the EU common methodology

[pic]

Report on the Pilot Phase (April – September 2005)

This report concerns the pilot phase launched in early April 2005 and aimed at testing ways of assessing administrative costs imposed by EU legislation. It provides the background of the pilot phase, its constraints as well as its main findings. The findings concern, on one hand, the methodology and, on the other hand, the technical capacity of potential data providers and their willingness to invest resources in the process.

The contents of the report have been kept as much as possible at a very factual level in order not to pre-empt the political decision of the College.

Background

The EC Treaty provides that the Commission should “take duly into account the need for any burden, whether financial or administrative, falling upon the Community, national governments, local authorities, economic operators and citizens to be minimised and proportionate to the objective to be achieved” (Point 9 - Protocol on the application of the principles of subsidiarity and proportionality). In order to comply with the proportionality principle, the Commission has therefore included the appraisal of administrative costs in its impact assessment guidelines and in its simplification efforts.

In its 16 March 2005 Communication on Better Regulation for Growth and Jobs, the Commission announced its intention to explore the possibility of developing a common approach for assessing administrative costs associated with existing and proposed Community legislation (COM(2005)97). The Staff Working Document annexed to the Communication outlined a possible common approach at EU level and announced the launch of a pilot phase to test the latter’s feasibility and value-added (SEC(2005)175). Results were scheduled for the autumn. The Commission also announced that, “after completion of the exploratory work, it will determine whether and how to integrate this aspect more fully in its standard methodology”.

On 23 March 2005, the European Council requested “the Commission and the Council to consider a common methodology for measuring administrative burdens with the aim of reaching an agreement by the end of 2005”. It stressed “that agreement should take advantage of the results of the Commission's pilot projects” and “that initiatives taken in the context of improving the regulatory environment must not themselves turn into administrative burdens.” (Pt 24, Presidency conclusions).

CONSTRAINTS

THE PILOT PHASE WAS CONFRONTED WITH FOUR TYPES OF CONSTRAINTS:

- limited time (in line with the Commission’s intention to make the fastest progress possible on preventing or reducing unnecessary burden, it was decided to set a very short deadline for the pilot phase – a bit more than 4 months leaving aside July and August, when many interlocutors are only partially available);

- broad policy scope (in line with the objective of developing a methodology covering all types of administrative obligations and all policy areas, it was decided to aim for a reasonably representative sample of cases);

- wide range of EU policy instruments (in line with the objective of covering the broad range of EU policy instruments, it was decided to test directives, regulations and strategic ‘packages’ – i.e. thematic strategies); and

- wide multi-level coverage (in line with the objective of delineating an “EU common methodology”, it was decided to test the capacity and willingness of the different levels of public authorities – EU institutions, national and sub-national authorities – and private organisations – European platform, national bodies and the addressees themselves).

The combination of these constraints was obviously quite challenging. It was therefore underlined from the start that the pilot phase might not be able to provide a final answer to all methodological and organisational questions. Thanks to the good will of Commission officials, their sectoral interlocutors at national level, the policy coordinating units of several Member States and the Directors and Experts on Better Regulation (DEBR) network, a lot of ground was eventually covered. Even so some questions remain to be addressed and some of the following findings need further confirmation.

METHODOLOGICAL FINDINGS

The presentation follows the logical order used in the Commission outline of the EU net administrative cost model (SEC(2005)175). The findings are based on the results of the pilot projects listed in Annex 2. These projects cover ex ante impact assessments and ex post evaluations of directives and regulations.

3.1 Aims and purposes of the model

The objective of developing a common model for assessing the costs of administrative obligations imposed not only on enterprises, but also on public authorities, the voluntary sector and citizens, was not challenged. Participants sometime implicitly recognised that this was in line with Treaty obligations. Moreover several Member States using variants of the Standard Cost Model (SCM) are already including citizens and the voluntary sector in their assessment.

The capacity of the model to assess administrative costs put on citizens was not tested because no EU measure in preparation or being reviewed for simplification between April and September would impose or imposes direct costs on citizens. It was reported that the general assessment of costs put on citizens in the Netherlands did not reveal any significant burden of EU origin[4]. The first step would therefore be to identify EU legislation imposing (direct) costs on citizens and see if that cost is substantial[5]. It would then be useful to see if putting a different value on the time required from EU citizens depending on their place of residence is politically too sensitive or if the crude but simple option described here above could also be used at EU level[6]. These points need to be further examined.

The objective of assessing net costs was accepted by contributors[7]. This was seen as consistent with Commission impact assessment guidelines and national RIA (Regulatory Impact Assessment) manuals; as well as being in line with the first OECD guiding principle for regulatory quality and performance. Moreover a net cost approach would have a clear advantage for those Member States that assess administrative burden systematically . First, with net figures, there is no need to go through costly periodical assessment of the entire legislation into force. Secondly, consolidated figures can be produced at any time which means that progress can be monitored on an ongoing basis (there is no need to wait for the general stocktaking exercise to know how total administrative burden evolved since the initial baseline measurement).

Despite the fact that the EU net administrative costs model is intended for microeconomic purposes (i.e. assessing specific pieces of legislation), the study showed that its standardised approach could also be useful for comparability and assessing cumulative burden[8].

Finally there was a very large consensus on the need to assess administrative costs in the wider context of a RIA (Regulatory Impact Assessment)[9] where costs and benefits can be put in perspective.

3.2 Definition of administrative costs

Terminology used at various levels and in different sectors proved to be very confused and confusing[10]. All parties recognised that there was no possible common EU methodology without a common definition setting very clearly the boundaries of administrative costs.

On the whole, the proposed definition appeared to be easy to comprehend[11]. Some however criticised the use of information obligations as the key element in the definition. They were often in favour of restricting information obligations to filling in forms, mainly because this is more in line with the conventional “red tape” connotation. The counter-argument was that filling in forms often requires gathering information first and that information gathering is often way more expensive than filling forms. Adopting such a narrow definition would therefore leave out many costly obligations and significantly underestimate actual administrative costs.

The reference to information obligations, also at the heart of the SCM definition, has the advantage of being comparatively straightforward and neat. It was indeed easy to distinguish between two types of legal obligations: some impose obligations on the behaviour or production process; others ask for information about that behaviour or production process. One service sustained that it was not always easy to distinguish between costs linked to policy design and costs induced by information obligations (cf. Marine thematic strategy). This point need to be further examined .

The disadvantage of referring to information obligations is that it leads to include actions that are not always intrinsically administrative (such as presenting a car for inspection in order to report on its conformity with safety requirements, or putting labels on manufactured products to inform consumers). It will therefore be important for analytical purposes to distinguish between the different types of information obligations. It should, for instance, be possible to disaggregate data into obligations to report to public authorities and obligations to inform consumers on products contents.

The need to take into account recurring costs and one-off costs was not contested. The study showed that many items initially identified as one-off costs were in fact recurring costs (equipment to be replaced at the end of the depreciation period). Actual one-off costs include the costs linked to the attestation of technical conformity and costs incurred when designing reporting and monitoring systems.

The prototype model proposed that administrative action required by law but corresponding to what an entity would normally do in the absence of any legal obligation should not be regarded as an administrative obligation. The ESTAT-led pilot project on the Intrastat regulation tends to show that a majority of businesses are capable of making that distinction, but with some difficulties. It is however not certain that this result can be extrapolated to other types of legislation and target groups. Considering the level of difficulty and the subjectivity which this distinction introduces, it seems preferable to drop it. A specific caveat underlining the fact that administrative obligations often overlap with ‘normal’ behaviour or managerial practices should however be added whenever appropriate.

The exclusion of administrative costs incurred as a result of participating in voluntary public programmes from the assessment was not seen as problematic. Some Member States however include such costs in their baseline measurement. It was therefore agreed that national databases should clearly distinguish between these ‘voluntary obligations’ and ‘compulsory obligations’. This would ensure sufficient comparability and additionality.

3.3 Core equation of the cost model

The main components of the core equation proved to be adequate. In all cases, the assessment could be based on the simple formula: average cost of an action (Price) multiplied by the total number of actions performed per year (Quantity). This was further confirmed by a review of previous analysis undertaken by the Commission (notably for the Impact Assessments of the REACH and INSPIRE proposals).

The pilot phase also highlighted that the price of an action cannot always be assessed on the basis of tariff (when work is done in-house, tariff = hourly labour costs + overheads) and time[12]. This is of course the case for equipment costs. Some also mentioned the fact that other variables could be used. For example, much of the analysis for the Marine Thematic Strategy, Soil Thematic Strategy, and the Urban Thematic Strategy was done by type-of-function. That choice does not prove that it is impossible to calculate the time taken for actions required by each function, but does suggest that time is not always the best analytical route to good estimates. It is also a reminder that for ex ante impact assessment of large policy packages setting very general obligations, it is much more difficult to go into detailed analysis. The level of details and accuracy should be adapted accordingly (see infra).

The pilot phase did test whether addressees were able to tell if there was a learning curve in meeting administrative obligations and if they could forecast further progress. Preliminary results showed that a significant minority can do so (questionnaire to the European Business Test Panel on the Intrastat regulation). That learning curve effect should therefore be taken into account whenever substantial. The simplest option seems to consider an initial one-off ‘learning/training cost’ in the first year alongside a recurring cost to carryout the administrative task each year (i.e. in the first year the overall costs would be higher). The learning effect could also be taken into account on the basis of a standard discounting ratio. That ratio could itself be defined by reference, for instance, to ICT productivity progress.

The study could not test whether the manner in which administrative costs are phased in (period of time and progressivity) needs to be taken into account and, if so, how. No suitable pilot project could be found. Comments made suggest that this dimension should only be taken into account on a case-by-case basis. In order to keep the model as simple as possible, this should be reserved for exceptional cases.

3.4 Scope and frequency of estimates

The study looked at the possibility of setting thresholds under which administrative costs do not need to be monetised. The underlying idea is to identify a simple variable that makes it possible to distinguish between what is substantial and what is marginal.

Using a threshold based on a minimum number of entities concerned by the measure seemed problematic insofar as some sectors are highly consolidated in Europe. It could therefore happen that measures affecting less than a handful of enterprises might overall represent very significant costs at EU level. The cost per action seemed equally problematic. What is financially marginal in one country, production sector, type of enterprises or group of population may vary greatly within the Union. As for the frequency, it does not provide in itself enough indication of the total burden.

For proposals of a restricted nature (that is, dealing with one specific dimension and setting relatively detailed obligations) and requiring little equipment, preliminary findings suggest that the number of hours imposed on each entity is the best option. The number of hours does not run into the problems described above and is probably the easiest component to assess. For equipment costs, a monetary threshold could be set on the basis of the lowest EU common denominator.

On the basis of the information collected, Denmark is the only country having set a lower threshold limit: laws imposing less than 100 hours of administrative work per year for all business concerned are not measured. Such limit seems however to be substantial for some target groups such as SMEs. What is to be considered as a negligible number of hours per entity therefore needs to be further tested.

Given the range of EU policies, it was suggested that thresholds should remain indicative. This would ensure that analysis will always be proportionate. The possibility of setting thresholds indicating what does not have to be quantified needs to be further examined.

3.5 Expected level of accuracy

All contributors share the view that t he level of accuracy should depend on the size of administrative costs induced by the (proposed) measure, the cost of the assessment and its potential benefits (proportionate analysis) as well as the maturity of the proposal. Proper implementation of existing rules on stakeholders’ consultation and quality control should be sufficient to guarantee against manifest error in determining the necessary level of accuracy.

The study confirmed that, when dealing with directives, it will often be difficult to be very precise when performing an ex ante assessment because usually the Member States only decide how to implement a directive after its adoption. There is however no reason to believe that assessing administrative costs on an ex ante basis will be more difficult or less accurate than the rest of the impact assessment. Broadly speaking the assessment of other compliance costs faces the same constraints.

The danger of spurious accuracy was underlined by Commission services[13]. In line with impact assessment guidelines, current estimates of administrative costs are usually expressed as a range. This reminds decision makers of the relative nature of the figures provided. The EU prototype – following the SCM in this respect – expects on the contrary a set figure for each action required by a regulatory obligation (see Reporting format below). Those contradictory requirements could, however, be reconciled relatively easily. Impact assessments should continue to refer to a range of estimates when presenting how costs have been assessed, but only use the median figure when summarising findings.

Preliminary results did not provide any indication on the average margin of error of administrative costs assessments. Apparently, Member States that contributed directly to the pilot phase, i.e. BE, DK, HU, NL, UK, have no data on this issue. Non participating Member States have since been invited to provide any evidence they may have. This kind of indication becomes particularly important when assessing cumulative burden (see below, last paragraph of ‘Available EU statistics’). One way to determine the average margin of error is to multiply ex-post evaluations of measures having been subjected to a detailed analytical quantitative ex-ante assessment. This point should be further examined.

3.6 Data sources

All contributors agree that the data sources listed hereafter can be used to assess administrative costs. In most cases, the assessment will have to combine different data sources. The relevance and reliability of each source must be asserted on a case-by-case basis. More specifically:

Available statistics in the EU

The study confirms that Eurostat is compiling a wealth of relevant data for the assessment of administrative burden, mostly on the number of entities (citizens or businesses) concerned at EU level and in each Member State; and on average wage per sector. The bookmarks ‘Periodical publications’ and ‘tables’ provide immediate access to preformatted information. Tailor-made tables can be produced through the ‘Data’ bookmark.[14]

Actual use of the databases for the pilot projects and random searches produced ambivalent results[15]. On one hand, it was possible to extract data fairly quickly (a couple of hours). On the other hand, it proved difficult to obtain information concerning the number of manufacturers at sub-sector level. The level of available information is sometimes so general that it was difficult to interpret its relevance. For example, it was difficult to determine the skill-set corresponding to the statistics on average labour costs. More importantly perhaps, for many countries, no sectoral data are available (this is particularly true for new Member States but the same problem occurs for many EU 15 countries); some time series do not go beyond 2001. This problem of compliance with existing statistical obligations could worsen given that some Member States are cutting down statistical work to save costs. With very few exceptions, it appeared very difficult for National Statistical Institutes to deliver on specific requests in a reasonably short time. This point should be further discussed.

The possibility of assessing administrative burden on the basis of existing EU statistics is therefore not a foregone conclusion. Adjustment and extrapolation will often be indispensable, all the more so if Member States continue to reduce resources allocated to statistical production. A number of existing weighting schemes and sampling techniques could help in that respect[16]. These schemes should ideally be listed. This point should be further discussed.

National databases on administrative costs could also be used. The Commission was informed of the existence of a database developed by EIM for the Dutch baseline measurement. A Danish database based on Microsoft Access has also been constructed. A UK database is in development. No technical information was exchanged at this stage. If this source of data is to be used to produce EU figures, it is of the utmost importance to ensure the interoperability of these databases as quickly as possible, to avoid heavy adjustment costs in the near future. This point should be further discussed .

All in all, whilst Eurostat databases are a valuable source of information, they will often not be sufficient by themselves. One must also recognise that the margin of error for EU figures based on this source will be rather high, because these figures will result from an addition of approximations and many extrapolations.

Standard ratios

The study concluded that it was desirable to establish standards for costs induced by standard information obligations. In order to simplify the assessment, it was also proposed to take overhead costs on a pro rata basis. Denmark and the Netherlands, for instance, usually consider that overheads amount to 25% of the labour costs. DG ENV opted for a 100% rate that seems more in tune with the type of work being undertaken in its area. The study did not manage to examine the possibility of setting discounting ratios for legal obligations corresponding to normal business operation. The development of a common methodology would require the harmonisation of these standard ratios. By definition, resorting to standard ratios would lower the assessment’s level of accuracy[17], but would also simplify greatly the assessment work. This point should be further discussed .

Opinion of experts

Not surprisingly, perhaps, the study confirmed that experts can be a very useful source of information and data for analyses. The main problem here was the need to communicate clearly enough to the experts the nature of the proposal so that they can feed in useful estimates in a timely manner. This problem exists for experts both within and outside Member State public administrations.

Data collected via targeted questionnaires

The study confirmed the difficulty of getting high response rates to targeted questionnaires either from national public administrations or private parties.

The pilot project on Intrastat also confirmed that designing a questionnaire, reaching agreement with services concerned and the European Business Test Panel national coordinators, and translating it into all Community languages was time and resource consuming. There should therefore be no obligation to include that source of data in all assessments. This does not mean that questionnaires should be automatically discarded either. Target groups in general and the business community in particular are often the only ones able to provide estimates on the time needed for specific activities.

The project on construction materials indicated that resorting to trade associations is sometimes a faster and more effective way of collecting data.

Field work and simulation

All agree that field work and simulation of the required administrative activities by independent experts should be the exception[18]. Contributors could not bring forward a concrete example of simulation done in the context of administrative costs assessment, but some promised to further investigate.

These methods could however be used as a last resort for appraising sensitive figures. For instance, businesses may be reluctant to indicate how fast they can meet an obligation because this would indicate to competitors where productivity gains can be easily achieved and ultimately dent their competitive edge. Another case in point would be the appraisal of costs in a new policy domain, where not even Member States have data and experience.

3.7 Division of responsibility

The basic assumption that ‘Member States are often best placed to conduct national fieldwork and simulations of typical national firm, association or citizen, while the European Commission is usually best placed to define what is needed and to aggregate data’ was not challenged. Participating Commission services indicated that, in some cases, they are only able to assess the upper bound costs, i.e. costs for Member States that have done nothing in the domain concerned. This can be in line with the principle of proportionate analysis.

However, when EU measures replace partially or totally national measures, the contribution of the Member States to assessing the actual additional costs imposed by EU legislation is often crucial. However, it was noted that as EU analysis is not typically undertaken on a country-by-country basis, the Commission services will always have an important role to play in extrapolating and interpreting the contributions of individual Member States.

On the relatively minor question of the ‘typical firm’, the most advanced users of the SCM[19] reckon that it should refer to the median firm in statistical terms rather than the average firm[20]. The pilot phase findings and past experience warns however against the risk of overall simplification if there are two types of typical firm (a large firm and a small firm). One can easily imagine an industry consisting of two or three large firms and a large number of very small firms. In such cases, merely giving consideration to the median firm (which will be a very small firm) would create the risk of miscalculating the overall cost to the industry. The best option seems therefore to assess costs according to two or more categories of firms size, whenever appropriate. This option should be further examined .

More importantly on the question of commitment to produce national data, participating Member States have clearly reiterated their previous pledge to contribute, but as yet they only represent a minority in the Union. As for the capacity to deliver, not surprisingly perhaps, the existence of the following seemed to bring a clear plus:

- a programme tackling administrative costs and endorsed at the highest political level, based on individual ministerial responsibility;

- a (central) coordination policy unit[21];

- a common methodology shared by all ministries

As a result, the three Member States meeting those three criteria (DK, NL, UK) made by far the biggest contribution to the pilot phase. Other Member States made very valuable contribution in the context of the DBR-led project. Out of 6 volunteer States (BE, DK, HU, IT, NL, UK), 1 resubmitted its initial national RIA (which was very good in the first place), 4 managed to produce an assessment and 3 of those used an analytical approach (i.e. providing figures for tariff and time required by specific actions). The Belgian and Hungarian assessments were particularly thorough. This result suggests that there is no major technical problem of feasibility at Member State level.

On the other hand, it must also be noted that the multiplication of fora dealing with administrative burden stretches the resources of central coordinating units.

At the same time, some of these units were reluctant to let Commission sectoral services interact directly with their counterparts. A similar reluctance was reported at the level of sectoral Ministries vis-à-vis their technical agencies that have a lot of the required technical information. This organisational dimension needs to be further studied to identify the fastest and cheapest way to communicate (technical) information between Member States and the Commission .

All in all, at the present moment, the number and distribution of contributors do not provide a sufficient basis for assessing costs at EU level (see Annex 3). A wider geographical distribution in particular appears necessary for all those cases that would require direct and specific information from grass-root level[22]. The test phase could not verify that a majority of Member States are politically committed and have the capacity to deliver. For specific areas, the willingness and capacity of subnational authorities (regions with legislative powers) should ideally be tested too. Some regional authorities manifested their interest but no suitable project could be found[23]. This point needs to be further examined .

3.8 Reporting

The study validated by and large the proposed reporting sheet. Results need to be reported in a standardised manner to allow them to be compared and added together. This being said, it should remain possible to provide additional information on the reporting sheet, as long as that does not alter the presentation of the report sheet’ standard part.

It was noted that, for proposals such as the thematic strategies, the common reporting sheet would often act as a summary of more detailed analyses. Services should therefore remain free to decide on the most appropriate format for communicating these detailed analyses.

In particular the study confirmed the need to include information on the transposition of EU obligations into national legislation (concordance table) and on differentiation between international and EU regulatory origin.

A number of possible solutions to technical problems were found. These include how to report on:

- equipment costs: these are not linked to human activity and cannot therefore be calculated on the basis of a tariff and time; in such cases, the ‘tariff’ and ‘time’ columns will be left empty; the yearly cost per equipment (i.e. based on the depreciation period) must be entered in the ‘price’ column; the number of machines needed by each entity concerned is reported under ‘frequency’; a standard depreciation method should be agreed; the straight-line method is probably the simplest option; costs of borrowing money to finance very expensive equipment could also be taken into account

- costs occurring every ‘x’ years: users should be invited to use the corresponding fraction (e.g. for biennial annual costs, ‘0,5’ should be entered in the frequency column)

- one off costs: a new column ‘one-off costs’ should be inserted next to the ‘target group’ column; putting a cross in this column would indicate that ‘1’ in the frequency column does not refer to a yearly frequency; alternatively ‘1’ in italics pr any special format could be entered in the frequency column.

- Costs not derived from the basis of time, would have to be reported simply under the cost column, and their derivation not shown.

The simplified typologies allowing quick reporting on the nature of the administrative obligation and on who will bear the costs were not tested at this stage. This is due to the limited number of pilot projects and to the fact that the impact assessments / simplification exercises concerned will only be completed in the coming months. This point should be further examined.

3.9 Methodological caveats

All contributors agree on the necessity of including caveats clearly drawing attention to the underlying assumptions and their effect on the accuracy of the assessment.

FEASIBILITY OF AN EU COMMON METHODOLOGY

On the basis of the findings presented in the previous sections, participants concluded positively on the feasibility of a common methodology built on the definition, core equation and reporting sheet of the EU net administrative costs model, as amended on the basis of the pilot phase findings and presented in the first part of this Staff Working Document.

The pilot phase also highlighted preconditions for the optimal introduction of an EU common methodology:

- adequate resources need to be made available at EU level; and

- a sufficient number of Member States coming from all parts of the Union need to contribute.

Lastly it confirmed that an EU common methodology would gain from clarification on a number of technical issues (see section 4.1). Optimisation efforts however did not appear as a prerequisite for the introduction of a common methodology. Embarking on learning by doing process would even help solving these issues.

4.1 List of methodological and technical issues to be further examined

Pending methodological issues are either minor or a matter of optimisation. Thanks to ongoing collaboration with Member States and the resulting methodological convergence, it should be possible to reach agreement with the Council on these issues quite rapidly. In any case, they are not an obstacle to the introduction of an EU common methodology.

Aim of the EU common methodology

1. Testing the capacity of the model to assess administrative costs put on citizens

Definition of the administrative burden

2. Full listing of borderline cases and how to deal with them

Scope of estimates

3. Validating appropriate thresholds indicating what does not have to be quantified (minimum thresholds).

Level of accuracy

4. Identifying the average margin of error of administrative cost assessments

Data sources

5. Testing the capacity of National Statistical Institutes to deliver on specific requests in a reasonably short time

6. Listing and, if necessary, developing weighting systems for assessing EU-wide costs on the basis of a limited quantity of national data

7. Ensuring the interoperability of national databases on administrative burden and access for the Commission

8. Harmonising standard ratios for overheads, training costs and learning curves and for costs corresponding to normal business operation, among other things.

Division of responsibility

9. Agreeing on the definition of a ‘typical firm’ or entity (median or average; cases where several categories of typical firms need to be distinguished)

10. Identifying the fastest and cheapest way to communicate (technical) information between Member States (including their regional authorities) and the Commission – organisational settings.

Reporting

11. Agreeing on simplified typologies on the nature of the administrative obligation and on who will bear the costs (target groups)

4.2 Costs of detailed quantification of administrative burden

4.2.1 Workload for Commission services

In the absence of national databases on the administrative costs generated by individual pieces of legislation, quantification of a single (proposed) legislative act at EU level based on the least demanding data collection methods[24] could require from Commission’s operational services or their external contractor between 14 to 40 hours of work over a period of 4 to 24 weeks.

Workload for the Commission operational service or contractor - Minimal configuration |

Tasks | Time (person hour) |

Reading the manual (one-off cost) | 2-3 |

Interface with the central policy / coordinating unit for further information | 1 |

Identifying the obligations of the (proposed) measure and required actions | 2 |

Searching main databases (Eurostat, etc.) | 2-3 |

Adaptation of a standard questionnaire to experts, national authorities, EU platform organisations or a few targeted entities, aimed at complementing database search | 1 |

Contact with the above by phone and via email | 4 |

Aggregating data, filling in and transmitting the report sheet | 2 |

TOTAL | 14-16 |

The workload for Commission central policy unit(s) would depend very much on the scope of application of the EU common methodology. The following table corresponds to small-scale application (assessment limited to a small proportion of (proposed) measures, no sectoral assessment, and no baseline measurement).

Workload for the Commission central policy unit(s) – Minimal configuration per year |

Tasks | Time (person hour) |

Following methodological developments and introducing updates | 160 |

Training | 200 |

External communication (brochures) | 160 |

Coaching of individual assessments | 400 |

Monitoring quality of individual assessments | 320 |

Reporting to EU institutions | 200 |

Exchanging best practices with Member States and international organisations | 160 |

TOTAL | 1600 |

Assessing major laws imposing many detailed reporting obligations would be much more costly, as analysis for REACH and INSPIRE has shown. This is confirmed by the cost of pilot measurements of national laws in Denmark and Norway: €100 000 for the VAT law or €100 000 for the Company Accounts Act.

4.2.2 Workload for national authorities

No information is available on the average cost or minimum cost of assessing individual pieces of legislation by public authorities. The overall cost of baseline measurement together with the number of acts measured would give an order of magnitude. Information on that number of acts has been requested from NL, DK and UK authorities. What follows is information on sectoral and overall costs of detailed and systematic quantification in the context of national reduction programmes.

In the Netherlands, the average baseline measurement per ministry was in the vicinity of €300 000. Each department with significant regulatory function has a team of 3 to 5 people dealing with the simplification / reduction programme. UK has the same staffing level.

The cost of the baseline measurement conducted in 2002 in the Netherlands is estimated to € 3 million (done by consultants). In Denmark the baseline measurement launched in autumn 2004 and due to be completed in October 2005 is said to cost €2 million, plus 4 full time persons over that period. Establishing the list of laws and regulations imposing administrative costs in the UK would require 6 to 8 consultants working full time for 3 months. Together with the baseline measurement, this is expected to cost around €26 million.

As for the overall implementation cost of national programmes aiming at 25% reduction of administrative burden over 5 years, the UK BRTF estimates that organisational costs for the reduction work will amount to c. €5 million per year for 5 years in the Netherlands compared with around €5,9 million in the UK. According to the same source, the Netherlands authorities have about 60 people dedicated to administrative burden reduction: (a) IPAL (Central administration) - 18 people; (b) Actal (Independent Oversight) - 9 people; and (c) Departmental Management Teams - 3-5 people in each department with a significant regulatory function.

4.2.3 Workload for private entities

Where contributors only have to compile existing data, the pilot phase showed that it could take between 2 person-hours (Intrastat regulation) and 24 person-hours (construction materials directive).

4.3 Contribution of Member States

For taking a decision on individual pieces of EU legislation, it did not appear necessary to have all Member States contributing. Nor did it appear necessary to demand that contributing Member States apply the EU common methodology to assess their (purely) national legislation. All the Commission would need from them would basically be to provide data in a standardised manner on the labour costs, time and number of operators affected by a (proposed) measure.

Evidence collected during the pilot phase clearly suggests that the involvement of national administrations will not be sufficient in a number of policy areas. Regional authorities also need to be involved and contribute.

ADDED VALUE / DOES ANALYTICAL QUANTIFICATION HELP MEETING BETTER REGULATION PRINCIPLES?

The proposed EU methodology has been found to be potentially useful in all pilot projects, provided there is scope for flexibility in the depth of the analysis. As already mentioned, the EU model must be able to cover a range of policy areas from employment to justice, but also a mix of policy instruments from strategic policies where administrative burdens are determined by Member States to detailed prescriptive legislation. The pilot phase helped to underpin three main advantages presented hereafter.

5.1 Analytical quantification helps in assessing measures from the point of view of those affected

Better regulation principles provide that it is i mportant to assess the impact of proposed measures and review existing regulation from the point of view of those affected. This is in particular indispensable for taking into account the distributional effects of a measure. It is an established practice in the Commission as for most contributors.

Most contributors to the pilot phase share the following reasoning. Having to analyse each obligation in order to determine what actions are required makes a difference. It obliges the drafter/reviewer to embark on a step-by-step simulation of what addressees will have to do. Because of the level of details required, the drafter/reviewer will usually have also to get in touch with a sample of those concerned and/or use experts to estimate the impacts on them. Just as with the SCM, the EU net administrative cost model provides an opportunity for the various parties to sit together. In turn simulation and data collection at grass root level help foreseeing different alternatives[25].

As major impact assessments and simplification exercises usually take more than 6 months, it was not possible in the limited time available for the pilot phase to verify how many alternative ideas quantification could bring in the policy design. This point needs to be further examined at the end of these exercises planned for the first half of 2006.

5.2 Analytical quantification contributes to transparency

Better regulation requires that public policy should be as transparent as possible. Reporting in a standardised manner on administrative costs facilitates the comparison between different options. It also facilitates the appraisal of the cumulative cost of the different measures applying to a sector, a profession or a group. This in turn contributes to improving regulatory transparency.

5.3 Analytical quantification provides an interesting indicator for monitoring progress and could help with the reduction of costs

Contributors agreed that quantitative analysis is potentially useful in helping Commission services to find ways to reduce administrative costs without endangering the wider objectives of proposals.

It is commonly accepted that, when objectives are quantified, they are more often reached (What gets measured gets done). Conversely, a general policy statement with no specific target and no performance indicator usually delivers little result. A quantitative approach would therefore help measure progress in the reduction of unnecessary administrative obligations imposed by individual acts or sectoral legislation.

It has also been underlined that when simplification efforts are quantified, they are more likely to be considered. Quantification would therefore help communicating on EU progress in this area.

Conclusions

The pilot phase showed that an EU common methodology is feasible and that it would bring added value. It identified a number of methodological points that could not be fully addressed in the course of the pilot phase and should be further examined. It also underlined various weaknesses concerning the availability and accuracy of basic data, largely linked to the problematic number and geographical distribution of Member States already able and willing to contribute. Finally it indicated that detailed quantification of administrative costs is a relatively expensive methodology that could not be conducted with the current level of staffing and financial resources available for assessment and evaluation.

Annex 1 – General and specific objectives of the test phase

The main objectives of the test phase were to test the feasibility and added-value of the proposed EU model and, where there is a range of methodological options, to assess each alternative. This was meant to cover, among other things, the specific issues listed hereafter:

12. the speed of delivery and operating budget required to produce estimates of administrative costs, in order to determine what can be reasonably delivered and define a proportionate approach;

13. specific problems with ex-ante assessment of administrative costs and with the evaluation of costs imposed on public authorities, the voluntary sector and citizens;

14. the accuracy of estimates produced in comparison with other methods (this would include verifying whether the model is applied consistently and how much the choice of data sources affects the results of the assessment, with particular attention to sampling techniques);

15. the value-added of the model in spotting unnecessary obligations ex-ante and ex-post, suggesting alternative solutions, and in setting priorities for simplification;

16. how to assess the regulatory origin of administrative costs (international, EU, national and regional law);

17. how to distinguish between major administrative costs and insignificant ones;

18. how to distinguish between what an actor would normally do for operational reasons and the additional burden due exclusively to legislation (cf. pricing, annual account or truckload inventory regulations);

19. how to take account of synergy effects and of the fact that time spent on each obligation declines over time with technological and human resource adaptations (this is important when there is no provision for a periodic general evaluation)

20. assess the complexity and subjectivity possibly added to the model by the obligation to distinguish between what an actor would normally do for operational reasons and the additional burden due exclusively to legislation;

21. examine problems posed by the inclusion of one-off costs in the definition of administrative burden;

22. see how to take depreciation into account.

Annex 2 - List of pilot projects

Type | Project Object / Title | Responsible service / organization |

Ex post | Improving knowledge on administrative costs imposed by Community Regulation (EC) No 638/2004 of the European Parliament and of the Council of 31 March 2004 on Community statistics relating to the trading of goods between Member States, in order to better manage these costs. Focus on the number of enterprises concerned, the modes of transmission of data at national level, number of hours required to prepare the data, evolution of the burden in time and usefulness of the statistics produced for the enterprises | European Commission - Eurostat |

Ex post | Contribution to the triennial review of Council Regulation 1158/2005 of 6 July 2005 amending Council Regulation 1165/98 of 19 May 1998 concerning short-term statistics (STS). Focus on reporting obligations on enterprises. | European Commission - Eurostat |

Ex post | Contribution to the preparation of the revision of Council Directive 89/106/EEC of 21 December 1988 on the approximation of laws, regulations and administrative provisions of the Member States relating to construction products. Focus on certification procedures and labelling obligations on enterprises. | European Commission - DG ENTR |

Ex ante | Methodological lessons from the appraisal of administrative costs made in the context of the impact assessment prepared for the Thematic strategy on air pollution (Commission's Work Plan 2005). Focus on the possible costs for Member States induced by the requirement to monitor PM2.5 concentrations[26]. | European Commission - DG ENV |

Ex ante | Methodological lessons from the appraisal of administrative costs made in the context of the impact assessment prepared for the Marine thematic strategy (Commission's Work Plan 2005). Focus in particular on the costs for public authorities to monitor Regional Marine Strategies. | European Commission - DG ENV |

Ex ante | Methodological lessons from the appraisal of administrative costs made in the context of the impact assessment prepared for the Pesticides Thematic strategy (Commission's Work Plan 2005). Focus on administrative costs for public authorities, pesticide users and pesticide producers induced by the collection of data on pesticides sales and use. | European Commission - DG ENV |

Ex ante | Methodological lessons from the appraisal of administrative costs made in the context of the impact assessment prepared for the Soil thematic strategy (Commission's Work Plan 2005). Focus on the costs of risk identification, including the preparation of an inventory of contaminated sites. | European Commission - DG ENV |

Ex ante | Contribution to the national impact assessment conducted by 6 Member States of the transposition of the Proposal for a Directive of the European Parliament and of the Council on the protection of groundwater against pollution, presented on 19 September 2003 (COM (2003) 550), following on the Water Framework Directive 2000/60/EC. Focus on monitoring and reporting costs put on public authorities. | Directors on Better Regulation (DBR) – Subgroup national RIA / Contribution from BE, DK, HU, NL & UK |

- | Methodological analysis of the evolution of the Standard Cost Model and possibility of adjustments to EU needs and resources. | European Commission – SG and DK, NL ad UK on behalf of the Informal SCM Network[27] |

Annex 3 - Overview of measurements for businesses per country (October 2005)

|NL |CZ |DK |DE |SE |EE |HU |PL |BE |IT |FR |UK | |Tax | | | | | | | | | | | | | |- VAT (excl. invoicing) |Y |P |Y |Y |Y |Y |Y |Y |Y |P |N |P | |- Excise Duties |Y |P |Y |Y |Y |N |N |N |P |N |N |P | |- Other Tax |Y |P |Y |Y |P |N |N |N |N |N |N |P | |Business and Econ. Affairs |Y |P |Y |N |P |N |N |N |Y |N |N |P | |Statistics |Y |P |Y |P |P |Y |P |N |P |N |N |P | |Justice |Y |P |Y |N |P |N |N |N |P |N |N |P | |Annual Reporting |Y |P |Y |Y |P |N |N |N |P |N |N |P | |Financial Markets |Y |P |Y |N |P |N |N |Y |N |N |N |P | |Environment |Y |P |Y |P |Y |N |N |N |P |N |N |P | |Business Permits |Y |P |Y |N |P |N |N |N |P |P |P |P | |Social Affairs |Y |P |Y |N |P |N |N |N |P |N |N |P | |Labour law |Y |P |Y |P |P |N |N |N |P |N |N |P | |Safety at Work |Y |P |Y |P |Y |P |N |N |P |N |N |P | |(Food) Safety |Y |P |Y |P |P |N |N |N |P |N |N |P | |Interior |Y |P |Y |N |P |N |N |N |P |N |N |P | |Health |Y |P |Y |N |Y |N |N |N |P |N |N |P | |Family, Consumer Affairs |Y |P |Y |N |P |N |N |N |P |N |N |P | |Science, Techn., Innovation |Y |P |Y |N |P |N |N |N |P |N |N |P | |Culture |Y |P |Y |N |P |N |N |N |P |N |N |P | |Transport |Y |P |Y |P |P |N |N |Y |P |N |N |P | |Y = Measured. N = Not measured. P = Planned to be measured or being measured

Source: Note from Dutch Ministry of Finance, UK Cabinet Office and Danish Commerce and Companies Agency for the SCM Informal Network, Differences in Application of the SCM - Analysis of the differences in application of the SCM , July 2005.

[1] Staff Working Document, ‘Minimising administrative costs imposed by legislation’, SEC(2005)175, 16 March 2005.

[2] This is but a provisional name for the future EU common approach. Its merit is to indicate clearly the nature and object of the model. Another option would be to build on the reference used by a significant number of Member States and use ‘EU Standard Cost Model’. The advantage of that option would be to build on a denomination already in use. The disadvantage would be to keep a fairly wide denomination referring in economics to any type of cost and therefore misleading for all those less versed in administrative burden matters. At domestic level, some users as Denmark are already using a different name in their national language.

[3] For compilation reasons, the same sequence of columns should be used to provide core information. Information required for national add-ons could be presented in the remaining columns

[4] Dutch Program on Reducing Administrative Burden for Citizens, Administrative burden for citizens , Ministry of Interior and Kingdom Relations / Services, Transparency and Administrative Burden Division, August 2005.

[5] Existing measures could be identified such as the EU Pet Passport but don’t seem to impose substantial burden.

[6] A note of the Belgian federal government on standardisation of tariffs released on 8 June 2005 provides interesting comparison on this issue.

[7] Net costs = costs introduced by a legislative act on minus the costs suppressed by that act at EU and/or national level.

[8] In the absence of specific guidelines, same sector projects initiated before the start of the pilot phase (Eurostat and ENV) have used different definitions of administrative costs and different reporting formats. As a result, it would have been difficult to compare and add the results of these individual projects.

[9] That point was in particular made in the Belgian contribution to the DBR-led project.

[10] Compliance costs are sometime used as a synonym for administrative costs, sometime as referring to all types of costs induced by legal obligations. Etc.

[11] Administrative costs are defined as the costs incurred by enterprises, the voluntary sector, public authorities and citizens in meeting legal obligations to provide information on their action or production, either to public authorities or to private parties. Information is to be taken in a broad sense, including costs of labelling, reporting, monitoring to provide the information and registration.

[12] Σ P x Q (Price = Tariff x Time; Q: Quantity = Number of businesses x Frequency).

[13] This point was particularly clear where the Commission is expected to foresee total administrative costs induced by implementation measures entirely defined by the Member States.

[14] http://epp.eurostat.cec.eu.int/portal/page?_pageid=1090,1&_dad=portal&_schema=PORTAL. See on the main page, under the heading ‘Themes’, links to ‘Industry, trade and services’, ‘Agriculture and fisheries, or ‘Population and social conditions’. For data on enterprises, see in particular the Structural Business Statistics (SBS) database with harmonised information for each EU country by industry (NACE classification) and size-class. Location: under the theme ‘Industry, trade and services’, the subfolder ‘Industry, trade and services - horizontal view’, and its subfolder ‘Structural Business Statistics (Industry, Construction, Trade and Services)’.

[15] In both cases, the focus was on business information.

[16] See for instance the 2002 ENSR Survey on European SMEs(http://europa.eu.int/comm/enterprise/enterprise_policy/analysis/doc/technicalnote_ensr_2002survey.pdf).

[17] Anecdotal evidence showed that IT and other equipment costs may vary significantly according to the required skills and the nature of work.

[18] Field work usually means conducting on-site or phone interviews with a sample of targeted entities asked to provide detailed information of the cost of each action. It could also mean one-site inspection, where the evaluator measures the work done in real-time (stopwatch approach). Simulation means that the evaluator will reproduce the work situation and perform the required actions to determine the time, level of skill and equipment necessary to fulfil information obligations.

[19] See Differences in Application of the SCM - Analysis of the differences in application of the SCM , Note from Dutch Ministry of Finance, UK Cabinet Office and Danish Commerce and Companies Agency for the SCM Informal Network, July 2005, p.4.

[20] The median is the middle value of a list, while the ‘ average’ – when it refers to the arithmetic mean – is the sum of a list of numbers, divided by the total number of numbers in the list. The reference to ‘average’ should in any case be avoided as potentially confusing. It is indeed indiscriminately used as a synonym for arithmetic mean, median (the middle value of a list), mode (the most frequent value) and weighted means.

[21] Where there is more than one leading ministry such as in the UK, things are by definition more time consuming but this additional workload was kept to a minimum thanks to good internal/national coordination.

[22] This does not imply that EU ambition should be to provide detailed estimates of net administrative costs for each Member State. In many cases, it would be disproportionately burdensome and subject to spurious accuracy.

[23] The impact assessment of the proposed groundwater directive gave the opportunity to the Flemish region to contribute.

[24] Rough estimate refers here to assessment requiring no formal public consultation open to all, no fieldwork and no simulation.

[25] See definitions under footnote 18.

[26] PM2.5: Particulate matter in ambient air with a diameter less than 2.5 millionths of a metre.

[27] The Informal Standard Cost Model Network regroups Member States leading or following the development of the Standard Cost Model. That group includes in particular the Dutch Ministry of Finance, the UK Cabinet Office & HM Treasury and the Danish Commerce and Companies Agency.