Development of Producer Price Indices for Services

Dipl.-Volkswirt Peter Roemer, Dipl.-Volkswirtin Astrid Stroh, Dipl.-Betriebswirtin Susanne Lorenz Development of Producer Price Indices for Services The system of official statistics still has considerable data gaps in the service sector. As regards price statistics, data deficits in the service sector predominantly concern the producer level. The information gaps are to be filled by the amended version of the EC Regulation concerning shortterm statistics of July 2005. The new Regulation provides for a quarterly supply of harmonised producer price indices within the European context on price base 2006 = 100 for selected top-selling industries of the service sector. This paper describes the background, goals and legal bases of the new producer price indices for services. It also deals with the development of the instruments used in price statistics. The pilot work on the development of producer price indices for the fields of legal activities, auditing activities, and tax consultancy is presented as an example. 1 Background and goals In most countries, Germany being no exception, the increasing importance of the service sector for the overall economy contrasts sharply with its statistical coverage. Based on a broadly defined concept of the service sector, the tertiary sector accounted for roughly 70% of the gross value added of all industries in Germany in 2004 according to national accounting data. 1 ) Growth is particularly high for business activities, for example in telecommunications, information technology, goods transport, business and management consultancy activities, advertising, legal activities, auditing activities, and tax consultancy. In the sectors of agricultural and industrial production in Germany, there are detailed and well-established sets of official statistics which have been compiled for a long time. Services provided for private consumption are also shown comprehensively in consumer price statistics and other statistical surveys. There are official statistics for the service industries of wholesale and retail trade, hotels and restaurants, too. Statistical coverage of business activities, however, started only several years ago. In 2001, the statistical offices in Germany conducted their first enterprise survey in the service sector, collecting data of the year 2000. 2 ) An improved picture of the economic strength, the structure and the economic development of the service sector 3 ) are the goals of that survey. Transport, communication services, real estate activities, renting of machinery and equipment and business activities are covered annually. In the field of price statistics, the service sector is not yet represented sufficiently especially at the producer level. Only a few countries, in particular the United States, Canada, Australia, Japan, the United Kingdom and France, have 1) As measured by gross value added in current prices; the services sector described includes trade, hotels and restaurants, transport, financial services, renting and business services, public service organisations. See Federal Statistical Office (ed.): Statistical Yearbook for the Federal Republic of Germany, 2005, Wiesbaden 2005, p. 631. 2) See Petrauschke, B.: First business survey in the services sector in WiSta 11/2002, p. 918 ff. 3) Included are sections I and K of the Statistical Classification of Economic Activities in the European Community (NACE Rev. 1). NACE is the abbreviation for Nomenclature statistique des activités économiques dans la Communauté européenne. 1248 Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005

so far made considerable efforts to cover producer prices in the service sector. Producer price indices are to measure the development of selling prices of goods and services provided by domestic producers. They serve to show the price development on the producer side, to show the business development, as deflators for measuring prices and volumes in national accounts (among others, to calculate the gross domestic product in real terms), as an indicator for the extent to which the goal of price stability has been achieved to the German Bundesbank at national level and to the European Central Bank at European level, and consequently for taking measures of monetary policy, as a basis of decision-making to the spheres of business and politics regarding action they have to take, to adjust recurring payments in business contracts (stable value clauses) on the part of businesses. It is necessary to develop producer price indices for services especially with a view to the requirements of national accounts. In accordance with ESA 1995 4 ), procedures with an insufficient basis of data may not be used to deflate production values in service industries. Product-specific producer price indices for services are regarded as the best method of deflating. As producer price indices are closely linked with national accounts in terms of subject matter, the development of methods needs to be coherent with national accounts and service statistics. That is the prerequisite for providing consistent and substantial overall data on the service sector. 2 Legal bases The Regulation of the European Union (EU) concerning shortterm statistics aims at establishing within Europe a common basic framework for the collection, compilation, transmission and evaluation of Community business statistics for the purpose of the analysis of the economic cycle. In accordance with the amendment to the EU Regulation concerning short-term statistics of July 2005 5 ), the Federal Statistical Office is obliged to supply, on a quarterly basis, producer price indices, harmonised in a European context, for selected priority 1 activities of the service sector using a price base of 2006 = 100. The EU Regulation concerning Short-Term Statistics attaches priority 1, first of all, to activities with high-volume turnover. Furthermore, the field of selected activities refers predominantly to those which so far have not been covered by other surveys of official statistics to a sufficiently large extent (e. g. consumer prices) or in respect of which there have been data deficits regarding the deflators of national accounts. The activities for which producer price indices are to be compiled are listed in scheme 1. Scheme 1: Activities 1 ) for which producer price indices need to be compiled NACE 1 ) Section I Transport, storage and communication NACE 1 ) Section K Real estate, renting and business activities No. Activity No. Activity 60.24 Freight transport by road 72.1 Hardware consultancy 72.2 Software consultancy and supply 61.1 Sea and coastal water 72.3 Data processing transport 72.4 Data base activities 62.1 Scheduled air transport 72.5 Maintenance and repair of 63.11 Cargo handling office, account ting and 63.12 Storage and warehousing computing machinery 64.11 National post activities 64.12 Courier activities other than national post activities 64.2 Telecommunications 72.6 Other computer related activities 74.11 Legal activities 74.12 Accounting, book-keeping and auditing activities; tax consultancy 74.13 Market research and public opinion polling 74.14 Business and management consultancy activities 74.2 Architectural and engineering activities and related technical consultancy 74.3 Technical testing and analysis 74.4 Advertising 74.5 Labour recruitment and provision of personnel 74.6 Investigation and security activities 74.7 Industrial cleaning 1) Statistical activity classification of the European Community, Rev. 1.1 (NACE Rev. 1.1, 2002). As can be seen from this list, the number of producer price indices to be newly developed is very big. Their development is an ambitious work programme, which we will only be able to master by making increased use of resources. This will be particularly difficult at times of widespread budget reductions. Development work is, however, substantively supported by a start-up funding granted by Eurostat, the Statistical Office of the European Communities, in the wake of pilot projects. As envisaged by the EU Regulation concerning short-term statistics, results with base year 2006 = 100 shall be supplied for all producer price indices from 2007 on. However, in accordance with a legal procedure transition periods ending in August 2008 may be granted for individual activities, 4) See European System of National Accounts (ESA) 1995 and Manual for Price and Volume Measurement in National Accounts. 5) Regulation (EC) No 1158/2005 of the European Parliament and of the Council of 6 July 2005 amending Council Regulation (EC) No 1165/98 concerning short-term statistics (Official Journal of the EU No. L 191, p. 1). Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005 1249

if the results are transmitted for the first time. Requests for transition periods must be substantiated by member countries. Eurostat will examine the requests and approve them, if they are well founded. Aspects related to the harmonised provision of EU-wide results by economic activity play a major role in that examination. To develop producer price indices in Germany a plan of several stages has been compiled, which comprises a number of years and relates with priority to certain kinds of activity. Pilot work started in 2003 on the development of producer price indices for activities such as telecommunications, legal services, auditing activities, and tax consultancy. This work has largely been finished by now. In 2004, pilot projects started in fields such as freight transport, post and courier activities, market research, business consultancy activities and advertising. They will probably be terminated in 2006. The kinds of activity to be addressed after that are architectural and engineering activities, technical testing and analysis, industrial cleaning, and investigation and security activities. The provision of personnel and the IT sector will conclude the chronology of development activities. The pilot surveys, conducted between 2003 and mid 2005 until the amendment to the EU Regulation concerning shortterm statistics No. 1158/2005, were voluntary surveys undertaken in compliance with article 6 of the Federal Statistics Law, which provides for measures for the preparation of surveys of federal statistics. Since August 2005, when the EU Regulation came into effect, respondents have been obliged to provide the information requested. 3 The development of price statistics tools At present, there are no fully harmonised concepts, neither nationally nor internationally, for surveys collecting data, which are needed to develop producer prices for services. This can mainly be explained by the following factors: Hitherto just a few states have gathered experience worldwide with the compilation of producer price indices for the services sector, the international exchange of experience in the field of price statistics started as late as in the 1990s, because of each country s specific national features, international experience can only in part be transferred from one country to another one, the price statistics methods to be used in the service sector are quite different from those used in the industrial sector, even within the service sector, price statistics methods vary between the different fields of activity. The approach presently practiced in developing producer price indices resembles to some kind of workshop. In most cases, the work done is geared to the experience that countries such as the United States, Canada, Australia, Japan, the United Kingdom and France have gathered in the compilation of producer price indices over the last 10 to 20 years. A pioneer role in the international discussion on methodological issues is played by the Voorburg Group 6 ). This group, founded in 1986, consists of service statisticians from more than 20 industrialised countries as well as representatives from the United Nations, the Organisation for Economic Cooperation and Development (OECD) and Eurostat. It provides an informal forum for the discussion of methodological issues of service statistics. It meets annually at alternating venues. Since the mid 1990s price statistics have increasingly come to the fore of the discussions. A lot of pioneer work on price statistics was internationally presented for the first time at one of the Voorburg meetings. In late 2002, the OECD and Eurostat established a joint Task Force, in which various EU countries, including Germany, as well as Australia, Japan and the United States participated and which was meant as a supplementary, rather than competitive organisation to the Voorburg Group. The goal of the Task Force was the development of a methodological manual on the compilation of producer price indices for services. This manual 7 ), which was discussed at several meetings, was completed in 2005 and published in spring 2006. The manual describes the pricing mechanisms usually practiced in the services sector broken down, as usual, by method of statistical price data collection, and it includes descriptions of methodological approaches in respect of the various fields of activity. The methodological manual distinguishes between pricing mechanisms, and pricing methods. Pricing mechanisms describe market processes and relationships, i.e. the ways in which the parties involved are estimating and negotiating prices on the market and how the services are charged to the customer s account. The term pricing method denotes the methods used by statistical offices to collect price data from respondents. The chart contains a list of the pricing methods most frequently used. Pricing mechanisms and pricing methods are closely related to each other in substance, but they are not identical. Exact knowledge of the pricing mechanisms in a given field of activity is a necessary condition to define a statistically adequate pricing method. Price data collection is restricted to major parts of the pricing mechanism and it assigns weights to the various pricing components. Below, various pricing methods, which are of importance to German price statistics, are explained in an exemplary way. 6) See http://stds.statcan.ca/english/voorburg/2004-index.htm (as of: 12 December 2005). 7) Methodological guide for developing producer price indices for services Edition 2005. This document can be downloaded both from the Eurostat and the OECD website. 1250 Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005

Main pricing methods Service output Unit of price measurement Clearly specified service Time spent providing services Type of price information Price observed Price estimated using related observed prices Price virtual Price observed or estimated Pricing method Direct use of prices of repeated services Unit value method Component pricing method Percentage fee method Model pricing method x) Time-based methods x) Note that the model pricing is classified entirely to the group of clearly specified services. Pricing where, as a part of price estimation, the number of required working hours is straight away assumed to remain the same as in the previous period, belongs to time-based methods. Joint OECD/Eurostat Task Force: Methodological guide for developing producer price indices for services Draft, 29/08/2005, http://www.stat.fi/voorburg2005/ppiguide.pdf, p. 28. 1 In the case of homogeneous and repeated services insofar as such data can directly be obtained from the respondent the methodological manual recommends that the best approach is the registration of transaction prices actually achieved for precisely specified services. Discounts and price reductions should be taken into account. 2 List prices that leave granted discounts out of consideration cannot, as a rule, serve as suitable variables for price index formation. They can be accepted as approximations to transaction prices only if they agree with market prices or if they change over time with the same change rate as market prices do. List or catalogue prices prevail in the private customers area. In contrast, the business customers area at least, transactions with major or corporate customers are dominated by special conditions, the evolution of which may be different from the development of list prices. 3 The billing of business services is based in most cases on the time needed for the work to be done, using hourly or daily rates. What is problematic here is the fact that the output is only measured by a time factor. A quality adjustment of the kind required by price statistics is not made, as technical progress and productivity efficiency are hardly liable to measurement procedures. From the point of view of data collection it is advantageous, as questions on time-based prices are easy to understand and can be answered quickly and easily. What is disadvantageous is the fact that the recorded hourly and daily rates or at least part of them represent prices which are offered, rather than actually achieved. As competition is getting tougher, the services sector, like other sectors, is increasingly granting discounts. Business consultants and market researchers, for example, do not reduce their hourly or daily rates, but work one or more days beyond the time originally scheduled without claiming additional payment. Such cases are difficult to identify, as enterprises are often not ready to lay open their transaction prices actually achieved. 4 Another important pricing method is referred to as percentage fees in the methodological manual. Thus, for example, commissions depend on the size of sales or fees are charged in accordance with the value of the subject of the legal work ( value of claim ). According to the methodological manual for the compilation of producer price indices for services, ad valorem prices used in Germany for tax and legal consultancy activities are assigned to the percentage fees charged as a proportion of the value of the subject of the legal work. As fees, which are charged as a direct proportion of the value of the subject of the legal work, prevail internationally and digressive fees are a specific feature of Germany alone, the term percentage fee was retained, although it cannot be taken for granted that the fee is always charged as a direct proportion of the value of claim. Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005 1251

This approach is a suitable method, in particular, in cases of legal and tax consultancy, where price data cannot be obtained by means of a direct interview, because the components of the services rendered are complex and adapted to the individual customer s needs and because they do not reiterate over time without changes. For more details, please, refer to chapter 4 below in this paper. 5 Finally, mention should be made of the unit price approach shown in the chart. It has been used in former pilot projects for the market research sector. The market research institutes had to indicate their prices offered and actually achieved per interview for surveys undertaken by interviewers (face to face) or by telephone. 6 Component pricing shown in the chart is an approach used in German pilot projects in the market research field for interviews based on a model case. The individual components project management activities such as planning and implementing the survey as well as the cost incurred by software programs which in most cases are not laid open to a customer are estimated as a total sum based on a model case in the survey and then they are recorded over time in price statistics. 4 Development of producer price indices in pilot projects of legal consultancy, tax consultancy and auditing 4.1 General approach to the project As mentioned above, projects for the compilation of services producer prices need to be implemented in a harmonised way at European level. However, it is necessary to take a country s specific situation into account, so that projects cannot be implemented in a standardised way. In addition, the diversity of the services sector makes it necessary to acquire special industry knowledge and to elaborate industry-specific concepts for price data collection. To start with, pilot projects, which serve to develop survey techniques and to prepare current price observation, should be launched for each industry in phases of experimentation. This work must be concluded first of all, before producer price indices by industry can be determined. The implementation of pilot projects and the compilation of price indices of official statistics are effected in two phases. The first thing to do is a formation of a basket of services with a corresponding weighting pattern. This is followed by current price observation. Basket of services and weighting pattern The weighting of products, fields of activity and/or billing modalities, based on the Laspeyres concept, is done according to the size of the annual turnover achieved by enterprises in the respective industry. Thus, the weighting is done in legal consulting according to the various fields of legal activity such as company law, tax law, traffic law, labour law, family and estate law, landlord and tenant law, property and building law, etc. as well as according to the billing modalities used such as billing based on freely negotiated prices and billing based on the legally prescribed scale of fees, in tax consultancy subdivided by services for income tax returns and business tax returns, business accounting, payroll accounting and for the provision of general tax advice as well as for extra outlays according to the Ordinance concerning the Remuneration of Tax Consultants (Steuerberatergebührenverordnung), in auditing both by activity, i. e. auditing as such, tax consultancy, business and management consultancy and by type of billing such as freely negotiated prices or prices based on the Ordinance concerning the Remuneration of Tax Consultants (Steuerberatergebührenverordnung). The weighting pattern is reviewed and adjusted at least once in five years to take account of market innovations and structural changes. A more frequent adjustment may be required in specific situations, for example in the case of scales of fees being changed. Current price observation The price representatives chosen for current price observation in line with the weighting pattern of a given industry should be typical of the industry concerned and achieve a high turnover in the respective sector. These market-specific prices and the observation of their evolution over time provide the basis for the estimation of producer price indices. 4.2 Pricing mechanisms of the German market The first prerequisite needed for the compilation of producer price indices is to have knowledge of the pricing techniques applied by those who are offering services on the market in fields such as legal and tax consulting or auditing to calculate the prices that will be charged to their customers for the services provided. Especially in comparison with Anglo-American countries it should be noted that the German market is legally regulated in many fields. The arguments of the legislator to limit the free play of market forces are manifold. This begins with a concern for consumer protection, as for example in the notary field, where free negotiation of prices between clients and notaries is not allowed at all and where notaries are even tenured in some federal states. In other cases as in the case of lawyers and tax consultants there are legally prescribed scales of fees for certain kinds of activity. However, 1252 Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005

in legal consulting there has been a clear tendency over the last few years towards freely negotiated prices. 8 ) Legal regulations exist for auditing activities as well, but they are less related to the pricing processes. As is shown by scheme 2, notaries charge their clients for all services on the basis of the notaries cost regulation (Kostenordnung). Lawyers, however, determine their fees both on the basis of a fixed scale (60%) and on the basis of freely negotiated prices (40%). Tax consultants, in their turn, charge their clients for 86% i. e. the major part of their services according to a fixed fee regulation. 14% of their fees are based on freely negotiated prices. Auditors bill all activities related to auditing exclusively on the basis of freely negotiated fees. Scheme 2: Billing modes Results of phase 1 of the pilot survey "Producer Prices of Services" of March 2004 Activity 1 ) as a... Fee freely negotiated % of total fee Scale of fees Denomination of the scale of fees Notary... 0 100 Cost regulation Lawyer... 40 60 Law on the remuneration of lawyers Tax consultant... 14 86 Ordinance concerning the remuneration of tax consultants Auditor... 100 0 none 1) Strictly speaking, a distinction is to be made between the bearer of an occupational group's name and the activity as such. Thus, for example, an auditor may also perform tax consulting activities, which can be billed according to the Ordinance concerning the Remuneration of Tax Consultants. 4.3 Compilation of weighting patterns The weighting pattern for each target group was defined in March 2004 by making a direct written inquiry among the various offices and enterprises. The inquiry was based on a sample per sector of economic activity stratified by size class of turnover. The sample was drawn from the business register under the maintenance of statistical offices in compliance with structural data of services statistics and recent information provided by experts. Offices and enterprises that achieved an annual turnover of more than 5 million Euros constituted a full-coverage stratum. Respondents were asked to reply to the survey for the weighting pattern on a voluntary basis, because the revised version of the EU Regulation concerning short-term statistics was not adopted yet by the time of the inquiry. The response rates vary between some 20 and 40% depending on the occupational group and have to be considered on the whole unsatisfactory. The weighting pattern assigns weights according to the turnover achieved in the various fields of activity. In a more detailed breakdown the fee and billing modalities, such as freely negotiated prices and/or fixed scales of fees, are additionally taken into consideration. Additional parameters were collected for the tax consulting sector. They will be needed later for the analysis of fixed scale fees in current price observations. The contents of the inquiry were adjusted with the support of the respective chambers and associations. Tables 1 3 show the distribution of turnover of lawyers, notaries, tax consultants and auditors by field of activity. This data can later be used to derive weighting patterns for the various target groups. Table 1: Distribution of turnover of lawyers and notaries by field of activity Results of phase 1 of the pilot survey "Producer Prices of Services" of March 2004 Percentages Share in 2002 turnover Billing type Field of activity total Fee freely Scale of fees negotiated Lawyers Company law... 18.0 15.3 2.7 Building law... 6.1 2.8 3.3 Labour law... 9.7 2.0 7.7 Landlord & tenant law.. 3.4 0.4 3.0 Traffic law... 3.7 0.1 3.6 Family and estate law.. 10.5 0.9 9.6 Tax law... 3.2 2.4 0.8 Other... 45.4 18.6 26.8 Total... 100 42.5 57.5 Notaries Family and estate law.. 19.4 0 19.4 Real estate law... 48.1 0 48.1 Company law... 17.6 0 17.6 Law of associations... 3.6 0 3.6 Other... 11.3 0 11.3 Total... 100 0 100 A question which auditors were additionally invited to answer in the inquiry for the weighting pattern was related to the time at which they were usually revising their hourly rates. As it turned out, most of the firms were changing their pricing rates at the end of the year. Tax consultants were interviewed in detail about their billing policy regarding common types of fees for tax assistance derived from the Ordinance concerning the Remuneration of Tax Consultants (StBGebV). This was done on the background that patterns or model cases were to be generated for future price observations, for which it was necessary to have concrete data on the amount of the fee. Discussions with the chambers have shown that changes in the level of fees defined by difficulty of work are accepted by the market in very rare cases only. That means that we can calculate with a value that remains stable over time, although it has to be checked in regular intervals, if this value is still valid. 8) In legal consulting the following approach to pricing is used: since the legal scale of fees for lawyers was amended by the introduction of the Lawyers Fees Law RVG (of 5 May 2004, Federal Law Bulletin I p. 717, 788) on 1 July 2004, legally fixed prices have had to be still applied in respect of the courts only. The prices for cases settled out of court have been freely negotiated with clients. Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005 1253

Table 2: Distribution of turnover of auditors by field of activity Results of phase 1 of the pilot survey "Producer Prices of Services" of March 2004 Percentages Field of activity total Share in 2002 turnover Billing type Hourly fee Flat fee with escape clause Scale of fees Auditing... 52.6 38.2 14.4 Tax consultancy... 28.3 14.4 4.6 9.3 Trustee activity... 0.5 0.1 0.4 Business & management consultancy... 16.1 12.1 4.0 Authorised expert/ evaluator... 1.2 0.7 0.3 0.2 Other... 1.3 0.8 0.4 0.1 Total... 100 66.3 24.1 9.6 A question was asked about the billing policy regarding fee (Rahmengebühren) ranges in line with 11 of the Ordinance concerning the Remuneration of Tax Consultants. The answers revealed different approaches depending on the field of activity. While the mean value seems to be preferred in business tax returns or book-keeping reports, Table 3: Distribution of turnover of tax consultants by field of activity Results of phase 1 of the pilot survey "Producer Prices of Services" of March 2004 Percentages Field of activity total Share in 2002 turnover Billing type Individual billing according to StBGebV 1 ) Flat fee according to StBGebV 1 ) Fee separately negotiated Assistance with tax matters Income tax returns... 14.3 13.6 0.3 0.4 Business tax returns, closings... 33.6 30.2 1.5 1.9 Business accounting: bookkeeping... 26.5 17.9 6.3 2.3 Payroll accounting... 9.6 6.9 1.6 1.1 General tax consulting.... 6.4 4.4 0.3 1.7 Representation in finance offices and courts, e.g. filing appeals, fines, tax criminal procedures.... 1.9 1.5 0.4 Extra outlays... 1.0 1.0 Services compatible with the profession of a tax consultant Business consulting... 3.8 3.8 Rating/ 18 KWG 2 )... 0.5 0.5 Trustee activity... 0.9 0.9 Investment counselling... 0.3 0.3 Expert opinions... 0.6 0.6 Other... 0.6 0.6 Total... 100 75.5 10.0 14.5 1) Ordinance concerning the remuneration of tax consultants. 2) German Banking Act (Kreditwesengesetz KWG). the focus of income tax returns is between the lower margin and the mean value. 9 ) Replies to a question about the most frequently used time fee (excluding VAT) (Zeitgebühr) show on average that the fees charged range between 30 and 46 Euros for one half hour. Fees charged for payroll accounting in line with fee ranges (excluding VAT) (Betragsrahmengebühr) according to 34 (2) of the Ordinance concerning the Remuneration of Tax Consultants amount in most cases to 10.00 or 12.00 Euros per employee and month, the sum billed on average amounting to 10.93 Euros. To display the typical tax consulting services we must build a system of model cases. The fee for a service depends, as a rule, on the value of claim. The value of the matter in litigation can be, for example, the sum of positive incomes in an income tax return or a requested remuneration of the capital gains tax in a capital tax statement. To build the model we must know the mean values as well as the lower and higher values of claim in their statistical distribution. Relevant information is available in some cases from secondary statistics. For example, important information could be obtained from the deeply specified data of income tax statistics of the Federal Statistical Office so that this field of activity was not included any longer in the survey. In other cases it was necessary to use the survey for the compilation of weighting patterns to collect statistical data on the values of claim. For example, data on the value of drafting financial statements or book-keeping accounts were obtained from tax consultants. In the survey respondents were asked to indicate major size classes (ranges) of values of claim used for billing most frequently were to be found. The analysis revealed relatively large ranges. In order to identify typical model cases, an analysis was made of the values most frequently occurring in the lower, medium and higher ranges. As a result, we built 16 model cases for tax returns of individuals, 38 model cases for business tax returns, 25 model cases for bookkeeping accounts of enterprises and 8 model cases for other kinds of activity. With a view to the correct timing of a possible later price observation, tax consultants were asked to provide information on the time or date at which their offices usually adjust the billing rates or recalculate the hourly or daily rates and as a result it was found out that in most cases the first quarter of a year is used to readjust the pricing rates. 4.4 Current price observation In accordance with the billing policy of a given target group, current price observation is subdivided into a first part comprising freely negotiated prices, which mainly consists of a question about hourly rates, and/or a second part comprising scale fees, showing the evolution of fees in line with the legally fixed scales on the basis of model cases, which are representative, but fictitious (see scheme 3). The model cases are directly analysed by the Federal Statistical Office without a personal interview. 9) The categories mean value, lower margin and upper margin are derived from the Ordinance concerning the Remuneration of Tax Consultants. The category to which a job is assigned depends on its difficulty; there is no discretionary scope for billing. 1254 Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005

Scheme 3: Methodological approaches to current price observation Field of activity Type of price Method Legal advice Scale of fees: Fixed rates Scale of fees: "Ad valorem" prices (according to the value of litigation) Hourly, daily rates A notary's activity Scale of fees: "Ad valorem" prices (according to the value of litigation) Tax consulting activity Scale of fees: Fixed rates Auditing activity Scale of fees: "Ad valorem" prices (according to the value of litigation) Hourly, daily rates Hourly, daily rates Concrete contracts Inquiries about the evolution of fees Analysis of the scales of fees Model approach for a "percentage (digressive) price progression" Survey by questionnaire Model approach for a "percentage (digressive) price progression" Model approach for a "percentage (digressive) price progression" Model approach for a "percentage (digressive) price progression" Survey by questionnaire Survey by questionnaire Direct interview at selected enterprises Between May and August 2004, lawyers and auditors were asked in a written inquiry to provide information about freely negotiated prices. The inquiry was updated in 2005. Notaries were not included in the inquiry, because all of their fees were based on the legally fixed scale. Since tax consultants charged most of their clients in line with the ordinance of fees for tax consultants, this group was not covered by the current price observation survey, either. The substance of the questionnaire consisted of questions about hourly rates. Law firms were asked to differentiate their hourly rates, if possible, according to the position held by the employee concerned, for example, secretariat, technical staff, junior or senior partner. Auditors were to differentiate their hourly rates according to four professional levels of skill and three fields of activity; in addition, there was a question on how travel costs were being charged. Strictly speaking, it would be necessary to ask auditors and lawyers on a quarterly basis about the amount of their hourly or daily rates, because the EU Regulation concerning short-term statistics provides for such periodicity. However, the information provided by the chambers concerned and discussions with experts as well as the outcome of the pilot survey have shown that the rates by the hour or by the day are adjusted once in a year or even less frequently. For that reason it seems justifiable for most enterprises to be covered, for the time being, by one inquiry in the course of one year; even more so as at the indicated moment of a change in the billing rates a questionnaire will be sent to the enterprises concerned. The response rate of the survey on the fees charged was somewhat higher (40%) than that of the survey for the definition of the weighting pattern, although we must assess the latter survey on the whole as less satisfactory, since the cases of non-response frequently related to enterprises with sometimes large volumes of turnover. In future it will be obligatory for respondents to provide the requested information. This will undoubtedly lead to an improvement in data quality. Model approach to percentage fees (degressive prices) The legally fixed scales include fixed rates and ad valorem prices. As fixed rates account but for a relatively small proportion of the total sales volume of lawyers, notaries and tax consultants, they are mentioned here only briefly. In most cases they refer to services of a routine nature such as, for example, the issuing of attestations. Changes in fixed rates and the related changes in prices can easily be identified by viewing the legally fixed scales of fees. For that reason there is no need for an inquiry. Ad valorem prices depend on the value of the subject of the legal work (value of claim). In Germany they are not typically proportional to that value; instead, a table compiled by the legislator assigns an exactly defined fee to an interval or range of that value. 10 ) The amount of the fees rises with increasing values of claim while the rates of increment are decreasing (degressive rating). To analyse the evolution of fees in the field of tax consulting it is necessary to include the underlying scale of fees (upper margin, mean value, lower margin) in the consideration, in addition to the value of claim. As described above, in order to undertake current price observations for that type of fees we identified representative cases for the various fields of activity, the specifications of which were kept constant over time. This ensures that the price statistical requirement of keeping the quality and the technical specification of goods and services constant over a certain period of time is fulfilled. Model cases were identified for lawyers, notaries and tax consultants, because the charges they bill to their clients are partly or completely based on legally fixed scales of fees. When defining model cases, we are faced with the difficulty of reconciling price statistics requirements with the need to identify cases with a substance that reflects reality as closely as possible. Problems are raised, above all, by legal consulting; for a specific legal case will never recur in the same way, neither in the course of time nor in respect of different clients. Thus, in order to construct such cases, experts have an intensive exchange of opinions, which is supplemented by literature research work. Model cases have the advantage that they can be analysed in a price 10) That means that with increasing values of a matter in litigation fees will rise in small leaps. For the fees to be statistically calculated on the basis of the value in litigation, a smoothing is done regarding the trend of the scale of fees. Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005 1255

statistical context without an inquiry. The disadvantage is that model cases represent fictitious, rather than real market transactions. As the definition of model cases is very expensive both for the system of official statistics and for the experts involved, the use of model cases has been restricted to those fields of activity that achieve significant turnover in terms of the weighting pattern. Price changes on the basis of changes in a legally fixed scale of fees are rare in Germany. The latest change in the scale of fees for lawyers occurred in July 2004 and the preceding one in 1994. The respective scale of fees for tax consultants was adjusted in 1998 for the last time. Thus, price changes related to a scale of fees are in most cases caused by changes in the value of claim. particular, of large enterprises with high turnover and for a refinement and further extension of model cases in some fields of activity. The presently used concept has the important advantage that respondents are troubled as little as possible. The rhythm of the inquiry is annual and the collection price data is partly done on the basis of model cases. This approach exempts notaries and tax consultants completely and lawyers partly from current price observations. However, the results will close an important gap in the statistical coverage of the services sector. In that way official German statistics meet an urgent demand of users, in particular, at European level. The amount of the value of claim changes over time. This may be caused, for example, by price changes and by fluctuations in turnover or incomes. For example, a change in real estate prices has an influence on a notary s fee for the notarisation of a contract of purchase and the amount of a tax consultant s fee depends on the changes in his or her clients incomes. What is problematic is the identification of adequate indicators which would enable us to extrapolate the price-related evolution of the value of claim over time. As a matter of principle, there should be a strong positive correlation between a reference indicator and the related value of claim. Indicators are items which are determined in the context of official statistics so that a reliable and consistent data base, which is continuously available, can be ensured. Reference indicators need not necessarily be price indices. If specific, subject-related indicators are lacking for model case computations, we can make use of general indicators, which show the evolution of a country s or sector s economic power (e. g. gross domestic product per inhabitant or the evolution of the sales volume or income per inhabitant). 11 ) On the basis of the weighting pattern, inquiries conducted and the evaluation of model cases we have defined the following provisional price indices with respective subindices for freely negotiated and scale prices: The results should be regarded as provisional, because there is still a need for an enlargement of the data base, in Table 4: Provisional price indices, 2004/2005 1 ) Fee freely Target group negotiated Scale of fees Total 2004 2005 2004 2005 2004 2005 Lawyer... 100.9 102.5 100.8 101.5 100.8 101.9 Notary... 100.9 102.4 100.9 102.4 Tax consultant... 100.3 101.0 100.3 101.0 Auditor... 101.1 102.7 100.3 101.0 101.1 102.5 1) Provisional base 2003 = 100. 11) A detailed description of the model cases for lawyers, notaries and tax consultants can be found under http://www.destatis.de/themen/d/thm_preise.php#pilotprojekt (as of: 12 December 2005). 1256 Federal Statistical Office Germany Wirtschaft und Statistik (Economy and Statistics) 12/2005