Content

Oliver Arentz, V.1 Services liberalisation in Germany: Overview and the potential of deregulation in:

Oliver Holtemöller (Ed.)

How Can We Boost Competition in the Services Sector?, page 177 - 240

1. Edition 2017, ISBN print: 978-3-8487-4676-7, ISBN online: 978-3-8452-8902-1, https://doi.org/10.5771/9783845289021-176

Bibliographic information
How to Assess the Economic Impact of Regulation – Recent Developments in Europe and Germany and Outlook to the Next Step Services liberalisation in Germany: Overview and the potential of deregulation72 (Oliver Arentz) Dr Oliver Arentz, Institute for Economic Policy at the University of Cologne Executive Summary The structure of the German economy changed drastically over time. The decline of the proportion of gross value added of the manufacturing sector at the expense of the services sector is often cited as an indicator for this structural change. However, this shift is not necessarily an indication of a decreasing importance of the manufacturing sector, but rather it points to a fundamental change of the gross value added process in manufacturing. Within the manufacturing process, business services and product-related services are growing in importance in development, production, marketing and sales as well as in customer relations and maintenance. With the increasing importance of intermediate inputs of the business services sector the interest in the regulatory framework for the provision of these services and the resulting welfare effect is also increasing. Inap‐ propriate regulation can, inter alia, lead to the waste of resources in the production process and to unexploited innovation potential. Negative ef‐ fects would especially occur in the downstream domestic production ar‐ eas, which compete internationally and, therefore, need competitive ser‐ vices. Conversely, the reduction of unnecessary regulation and trading- V V.1 72 Arentz, O.; Manner, H.; Münstermann, L.; Recker, C.; Roth, S.J. (2015): Services liberalisation in Germany: Overview and the potential of deregulation, Otto-Wolff- Discussion Paper, No. 01b/2015. 177 barriers can release unused growth potential not just in the services sector, but also in the manufacturing sector. The OECD has recorded the level of regulation for selected profession‐ al services since 1998. The indicators of the OECD confirm that substan‐ tial structural reforms have taken place in Germany in the past. Nonethe‐ less, Germany does not have one of the lowest indicator values of EU countries. In addition, other regulation measures like the Service Trade Restrictiveness Indicator also published by the OECD and the level of economic rents suggest further deregulation potential in the professional services sector.This study offers three principles for the identification of potential deregulation targets. First, we recommend the equalisation of dif‐ ferent profession-specific regulations for professions with comparable fields of activity to the currently lowest level of regulation. Second, one can consider the equalisation of state specific regulations within Germany for the same professions to the lowest existing intensity of intervention. Third, ‘gold-plating’ of existing and future EU demands should be avoid‐ ed. The results of the econometric analyses for selected business services carried out as a part of this study confirms the significant impacts on wel‐ fare of past structural reforms on the net value added in the downstream manufacturing production as well as in the overall economy. Furthermore, the discovered interrelationships give rise to positive indications for the expected effects of potential future reforms, which should have growth ef‐ fects particular on the German manufacturing sector. Introduction More than 7 out of 10 employees in Germany work in the services sector. In particular, the business services sector grew dynamically over recent years. Moreover, industrial products include an increasing proportion of services; the division between industry and services is becoming less clear-cut. This can also be shown in the employment structure of the man‐ ufacturing sector, where an increasing amount of employees provide ser‐ vices. Services and industrial products are closely interconnected and de‐ pendent on each other. Since the beginning of the 2000s the EU has in‐ creasingly focused on the services sector. The directive on services in the internal market and the directive on the recognition of professional qualifi‐ cations are cornerstones for the development of the internal market in the 1 V How to Assess the Economic Impact of Regulation 178 services sector. Current efforts focus on reducing market barriers and on thereby creating positive welfare effects. In Germany the debate on services liberalisation is apparently observed rather than actively advanced. Many proposals from Brussels lead to resis‐ tance from lobby groups. But there appears to be room for competitionenhancing deregulation as this study shows. Initially we define the term “service” and give an overview over the German market (chapter 2). In chapter 3 we resume the debate regarding the alleged services gap in Ger‐ many and scrutinise it by further analysis. Next we give an overview over the status quo of regulations in the services sector (chapter 4) and of cur‐ rent steps for services liberalisation (chapter 5). Afterwards we analyse the economic effects of services liberalisation both on a theoretical as well as on an empirical level (chapter 6). In chapter 7 we outline which politicoeconomic resistance against further steps for services liberalisation is to be expected. Finally, we summarise the results and show steps for economi‐ cally reasonable deregulation, which do not seem politically hopeless (chapter 8). The German services market – definition and overview What exactly are services? In the literature the question of how to clearly separate manufacturing and services is intensively discussed. A widespread definition characterises services in contrast to manufacturing as non-material processes. According to this definition, the goal of the ser‐ vices process is to change the condition of a product or a person (Hill 1977). Following this process thinking, services, in contrast to products, cannot be produced in advance as they are not storable until their sale or usage. Rather the processes of creation and of usage have to take place si‐ multaneously. Many services additionally require the presence of the consumer and the interaction with the provider. Examples are consumer-related services like being examined by a doctor, riding a taxi or attending a concert. These services are called time- and place-bound services, as their creation is linked to location and time. Due to the personal nature of the work, it is usually assumed that there is only a minor potential for development or in‐ creases in productivity (Eickelpasch 2012). Another consequence of the necessary simultaneous presence of the services provider and the con‐ 2 V.1 Services liberalisation in Germany 179 sumer is that the tradability of these services over a greater distance is very limited. Due to the appearance and the further development of modern commu‐ nication technologies this simultaneous presence is necessary for fewer and fewer services. Many services become unbound services. Examples are the remote maintenance of equipment or hardware, data teleprocessing or many financial services (Eickelpasch 2012). In contrast to time- and place-bound services, unbound services just like manufactured products are usually tradable over a greater distance. Moreover, in many cases the creation of a service does not have to take place simultaneously to its us‐ age. Some unbound services like manufactured products can therefore be created in advance. For example, the remote maintenance of equipment or technical planning and calculations for a customer can in parts make use of standard components from a provider’s own product portfolio. In the case of unbounded services, increases of productivity similar to the in‐ creases in manufacturing are possible. These increases can mainly be evoked by economies of scale. This short overview shows that a clear dif‐ ferentiation of services and manufacturing among themselves is becoming more and more difficult. For the following analysis of the services sector a functional view seems appropriate: The services and products in the services sector can be classified by considering whether they are meant for final consumption or as an intermediate input for manufacturing or other services (see Figure 5.1.1). Services sectors which are predominantly used as intermediate in‐ puts, are called business services. Examples are advertisement and market research, consultancies, external research providers or whole sale. Busi‐ ness services can be separated from services in tourism, leisure, retail or the public sector, which are created predominantly for the final consump‐ tion and which are part of consumer-related services. V How to Assess the Economic Impact of Regulation 180 Proportional use of domestic services Notes: Last usage of domestic services products, intermediate inputs include manufac‐ turing investments, assumption: domestic usage structure for exported services prod‐ ucts; year 2010 Source: input-output calculations of the German Federal Statistical Office, own calcu‐ lations and illustration. In the context of structural changes of the economy, services have grown in significance in Germany (see Figure 5.1.2). Following the widespread Figure 5.1.1: V.1 Services liberalisation in Germany 181 sectoral partition in a primary sector of initial production, a secondary or manufacturing sector and a tertiary or services sector, about 69 per cent of the domestic gross value is contributed by services. The share of employ‐ ment is even slightly higher with 74 per cent. The manufacturing sector contributes about 26 per cent of gross domestic value and about 19 per cent of employment. However, many occupations in the manufacturing sector today concentrate on services. Therefore, the overall importance of services cannot fully be comprehended by considering the sectoral econo‐ mic structure (see chapter 3). The construction sector is statistically count‐ ed as part of the manufacturing industry and contributes about 5 % to do‐ mestic value and to employment respectively. Share of gross value added by economic sectors in Ger‐ many Note: Before year 1991: former West-Germany. Source: German Federal Statistical Office, own draw based on Eickelpasch (2012). The increase in importance of the services sector is partly due to the in‐ crease of public services e.g. in health care or the social sector. More im‐ portantly, the increase in business services contributed to the increase of the services sector. There is an increasing demand for business input fac‐ tors by other services providers as well as manufacturers accompanying products within their production or the export of products. Since the Ger‐ man reunification in 1991 both the gross value added and the employment Figure 5.1.2: V How to Assess the Economic Impact of Regulation 182 in the area of business services grew the most. Business services firstly consist of knowledge-intense services like legal services, consulting, tax and public accounting, architecture, engineering as well as advertising and market research. Secondly also less knowledge-intense services like secu‐ rity services or rental and leasing activities are counted as business ser‐ vices. Lower increases can be seen in information and communication ser‐ vices or predominantly consumer-related other services. However, the pro‐ portion of financial and insurance services as well as retail, hotel and guest industry and transportation within the domestic economic structure have almost not changed (see Figure 5.1.3). Increase of the gross value added and of employment in the domestic services sectors Note: Time period 1991 until 2013, in per cent, gross value added inflation-adjusted for base year 2010 Source: German Federal Statistical Office, own illustration. There are considerable differences regarding the profitability within the area of business services. The profit to revenue ratio is usually used as a figure of the profitability of a company. In the national input-output calcu‐ lation, the respective sector-specific net operating surplus is reported. The net operating surplus is the operational result after the subtraction of inter‐ mediate inputs, write-offs as well as expenditure on the compensation of employees. For independent companies the net operating result is the mixed income before the subtraction of taxes. Whereas the proportions of Figure 5.1.3: V.1 Services liberalisation in Germany 183 the gross value73 of the output value are slightly below 50 per cent both in knowledge-intense as well as in less knowledge-intense services, consider‐ able differences regarding the net operating surpluses do exist. The sector of knowledge-intense business services made an average net operating surplus of 16 per cent of the production value in 2010 (see Figure 5.1.4). Particularly high surpluses were achieved in the areas of marketing and market research, legal services, consulting as well as architecture and en‐ gineering. In less knowledge-intense services the net operating surplus was considerably lower, namely 6 per cent on average. In the domestic in‐ dustry the net operating surplus was on average 9 per cent in 2010. Gross value added and net operating surplus in the area of business services Notes: Year 2010, in per cent proportional to production value Less knowledge-intense services: retail, transportation, warehousing, renting and leas‐ ing, hiring-out of workers, traveling agencies and events, economic services (among others security). Knowledge-intense services: media and publishing, computer and information ser‐ vices, telecommunication services, financial and insurance services, legal, tax account‐ Figure 5.1.4: 73 The gross value added equals the production value minus the intermediate inputs. After subtracting write-offs you have the net value added. The relevant difference to the net operating surplus is that the compensation for employees is included in the gross value added, whereas the net operating value is the part of the gross val‐ ue added, which a company has at its disposal for investments, capital service, tax‐ es and withdrawals of profit. V How to Assess the Economic Impact of Regulation 184 ing and consultancy services, architecture and engineering, research and development, other liberal professions. Source: Input-output calculations of the German Federal Statistical Office, own calcu‐ lations and illustration. The domestic market is by far the most important market area for services created in Germany. Although the proportion of the services sector of the domestic gross value added is three times larger than the proportion of the manufacturing sector, 85 per cent of all German exports are industrial products. The share of exports in consumer-related services (without pub‐ lic services) like culture and leisure, retail or personal services like hair‐ dressers or laundromats were below 4 per cent in 2010. The main reason is assumed to be the previously mentioned limited tradability of these ser‐ vices, which are linked to the provider. The proportion of exports of busi‐ ness services was slightly higher. On average less than 17 per cent of do‐ mestically created less knowledge-intense services were exported (see Figure 5.1.5). Within the less knowledge-intense services there are, how‐ ever, huge differences. While wholesale and transportation show above average export quotas, place-bound services like security were almost ex‐ clusively created for the domestic market. The services and products of knowledge-intense services providers are increasingly part of the unbound services. Although services provider and customer do not have to be present permanently during the creation process, the export ratio of knowl‐ edge-intense services is only 12 per cent. Within the knowledge-intense services, the areas research and development, information and communi‐ cation services as well as architecture and engineering show an export ra‐ tio above average. The volume of services exported to the rest of the world surmounts the export volume to members of the European Union in many services sec‐ tors. Worth mentioning is the comparison with the entirety of all German goods and services exported in 2010. With a proportion of roughly 60 per cent, there were more exports to member states of the European Union than to the rest of the world. V.1 Services liberalisation in Germany 185 Export ratio in the sector of business services Note: Year 2010, in per cent proportional to the total usage goods in a sector Source: Input-output calculations of the German Federal Statistical Office, own calcu‐ lations and illustration. Services gap – the relationship between the services sector and the manufacturing sector In the debate about the German economic structure it is regularly re‐ marked that Germany has a so called services gap, which could endanger the international competitiveness of the German economy in the long run. The discussion on the alleged services gap is based on the observation that in international comparison, the manufacturing sector still makes up a high proportion of the German economy. In other European economies like France or the United Kingdom the proportion of the manufacturing sector has decreased significantly in recent years, the manufacturing sector cur‐ rently only contributes 13 and 16 per cent of their gross value respectively. In Germany the manufacturing sector’s proportion of the gross value has remained almost constant during the same time period. With an share of manufacturing of about 24 per cent Germany is well above-average com‐ pared to the economies of the European Union or the US (see Figure 5.1.6). Figure 5.1.5: 3 V How to Assess the Economic Impact of Regulation 186 The manufacturing sectors proportion of the gross value added by international comparison Notes: In per cent, years 2000 and 2010; manufacturing sector without the construction sector Source: OECD, own illustration. Sometimes the comparatively large share of the manufacturing sector in the German economy is viewed with criticism. An argument commonly brought forward says that a high share of manufacturing of the gross value added identifies a structural weakness of an economy, since a transition to a modern services society apparently failed74. This judgement is based on the idea that the average gross value added by manufacturing is lower than in services provision, which is particularly disadvantageous for employ‐ ment in countries with a high wage level. The reason is primarily that sim‐ pler jobs with a low gross value added in manufacturing are affected by automation processes or relocations. Simpler services jobs are usually consumer-related and time-related tasks and are therefore barely affected by global relocations. However, simpler services are affected by automa‐ tion processes and rationalisation processes. Examples are the disappear‐ Figure 5.1.6: 74 See the description by Schmidt (2012), whereas the author does not adopt this pos‐ ition. V.1 Services liberalisation in Germany 187 ance of former services jobs like the petrol attendant or the appearance of self-service cash points in super markets. In the debate about Germany’s high share of manufacturing and the al‐ leged services gap the structure of the domestic industry is partially mis‐ judged. Sometimes the notion that the jobs in the manufacturing sector are mostly limited to physical manufacturing and therefore traditional jobs in manufacturing seems to be dominant. Nevertheless, within the manufac‐ turing sector a structural change to servitisation has occurred that is simi‐ lar to overall structural change of the three sectors of the economy. Within the domestic manufacturing sector employment in fabrication decreased for several years, while employment in product-related services like re‐ search and development or customer services are increasing continuously. For an industrial product to succeed in the market it is often not sufficient to solely produce goods. Customers increasingly expect customised solu‐ tions to their problems, which means that product-related services play an increasing role in the value of the product (Eickelpasch 2014, De Baker et al. 2013). This part is in many cases the decisive competitive advantage. Industrial companies are therefore faced with the question whether they want to provide product or sales-related services like development activi‐ ties, sales, customer services or legal services and strategy advice them‐ selves or whether they want to purchase these services from third-party providers. With the currently used statistical classification, the statistically recorded total size of the services sector and manufacturing sector in a country depends on which alternative industrial companies choose (Döhrn et al. 2008). A differentiated view on the services share within the manufacturing sector and therefore on the total economic structure is offered by the em‐ ployment statistic of the German Federal Labour Market Authority. The statistic discloses the primary field of activity for every employee. In this way service tasks within the manufacturing sector can be well recorded (Eickelpasch 2014). In 2013 almost 50 per cent of all employees in the manufacturing sector executed service tasks (see Figure 5.1.6). The largest share of these service tasks were found in the areas internal research and development as well as management and organisation of a company. The third largest area of internally provided services encompasses tasks in ac‐ quisition and sale, in customer services as well as in marketing and prod‐ uct placement. Moreover, in many manufacturing sectors in-house logis‐ tics providers are used. In some export-orientated manufacturing sectors the proportion of services employees is well above the average of the V How to Assess the Economic Impact of Regulation 188 overall manufacturing sector. In the domestic pharmaceutical industry 9 out of 10 employees are services employees, in the chemical industry it is 8 out of 10 employees. In particular, these two examples show how mis‐ leading the statistical classification between the secondary or manufactur‐ ing sector and the tertiary or services sector can be. Services sector employees in the German manufacturing sector Note: Proportion of total employment, year 2013 Source: Employment statistics of the Federal Employment Agency, own calculations based on Eickelpasch (2014), own illustration. Döhrn et al. (2008) assume that industrial companies in the Anglo-Saxon area are more likely to use external services suppliers. German industrial companies rather prefer to control more operations themselves by keeping them in-house. These differences can partly be explained by different cul‐ tural traditions and corporate philosophies, diverging considerations for costs and efficiency as well as the need to prevent external parties insight into the company. However, the hypothesis that industrial companies are unable to find an adequate supply of intermediate inputs for some of the required services and therefore use in-house solutions is also reasonable. Figure 5.1.7: V.1 Services liberalisation in Germany 189 Despite the comparably large services share of the domestic industry, purchased services as intermediate inputs also continuously increase in importance. In 2010 the industrial sector used intermediate inputs in the amount of 1,385 billion Euros. With 368 billion Euros about a quarter of these intermediate inputs came from the services sector (see Figure 5.1.8). Furthermore, a part of the intermediate inputs bought by industrial com‐ panies from other industrial companies are also services. With the produc‐ tion of accompanying services many industrial companies gain specific skills which they not only use for internal purposes. Rather many industri‐ al companies directly appear as services providers themselves and offer their specific skills to third parties. Intermediate inputs purchased by the manufacturing sector Note: Year 2010, in billion Euros Source: Input-output calculations of the German Federal Statistical Office, own illus‐ tration. While the structure of industry-internal input services cannot be measured in detail, the input-output calculations of the German Federal Statistical Office offer a differentiated overview over the intermediate inputs from the services sector. Considering the volume of purchased business services from the services sector, one can see that wholesale has the highest impor‐ tance for the manufacturing sector. The second most important are legal services, tax and public accounting and consulting before transport and lo‐ Figure 5.1.8: V How to Assess the Economic Impact of Regulation 190 gistics. Architecture and engineering firms, other economic services like security services, rental and leasing, labour recruitment and provision of personnel as well as marketing and market research are also of high im‐ portance. The importance of external research and development activities varies significantly between different manufacturing sectors, whereas the industry average is negligible (see Figure 5.1.9). Business services purchase by the manufacturing sector Note: Year 2010, in billion Euros Source: input-output calculations of the German Federal Statistical Office, own calcu‐ lations and illustration. On the basis of the open statistical input-output model of the German Fed‐ eral Statistical Office and thereby considering the external trade relations, Figure 5.1.9: V.1 Services liberalisation in Germany 191 one can understand how many units of a good of a domestic services cre‐ ation are directly and indirectly needed to provide one unit of an industrial good for final usage (Schmidt 2012). The intermediate inputs from the ser‐ vices sector are therefore of particular importance for export-orientated manufacturing sectors like the chemical industry, the construction and me‐ chanical engineering sector as well as the automotive industry (see Figure 5.1.10). For example, 0.43 units of intermediate inputs from the services sector are needed if the production value increases by one unit on a value basis. Value based proportion of intermediate inputs from the services sector for the production of another value unit of domestic manufacturing production Note: Year 2007, on the basis of the statistical input-output-model of the German Fed‐ eral Statistical Office Source: Inverse coefficients of the input-output calculations of the German Federal Statistical Office, own illustration based on Schmidt (2012) On the one hand, as the manufacturing sector is an important recipient of services form the services sector, this demand creates direct growth and employment effects in the services sector. Assuming the services demand‐ ed by the manufacturing sector go along with an identical employment share in the services sector, the demand for industrial products caused em‐ ployment in the amount of 3.8 million employees in the domestic services sector in 2009 (Eickelpasch 2013). Figure 5.1.10: V How to Assess the Economic Impact of Regulation 192 On the other hand, the effects of the services sector in the manufactur‐ ing sector are less direct. The demand for industrial intermediate inputs used for services creation amounted to only 20 per cent of the overall in‐ termediate inputs of 1,084 billion Euros and was, therefore, lower than vice versa, as the services sector receives the major part of its intermediate inputs from other services providers (see Figure 5.1.11). Due to the pur‐ chasing of industrial intermediate inputs by the services sector direct de‐ mand effects, nevertheless, do exist. Furthermore, services intermediate inputs increasingly contribute to the development as well as the produc‐ tion and the distribution process of industrial companies. This suggests that the services sector does not only create demand effects, but also direct effects on the manufacturing sector. Both improvements in quality as well as in cost saving can, for example, positively contribute to the gross value added of the manufacturing sector during the course of deregulation ef‐ forts and the reduction of barriers of trade. A discussion of the supply ef‐ fects of business services on the manufacturing sector can be found in sec‐ tion 6.2. Intermediate inputs purchased by the services sector Note: Year 2010, in billion Euros Source: Input-output calculations of the German Federal Statistical Office, own illus‐ tration. Figure 5.1.11: V.1 Services liberalisation in Germany 193 Empirical survey of the status quo of the regulation in the services sector The regulatory intensity in the services sector can be derived from the to‐ tality of all specific regulatory interventions in the relevant legislation and decrees. An actual evaluation of all relevant sources is almost impossible. Particularly for international comparisons, this procedure would have to be undertaken for all countries. Therefore, empirical research usually re‐ sorts to existing regulatory indicators. The OECD has been measuring the regulations in the non-manufactur‐ ing economy on a sectoral basis in 34 OECD and 22 non-OECD countries every five years since 1998 (OECD 2013, Product Market Regulation Database, www.oecd.org/economy/pmr). Apart from the indicators for network sectors (energy, transportation and communication) and retail the OECD offers an indicator for “professional services”, which covers the professions public and tax accounting, legal, engineering and architecture. The “professional services” are of particular importance as a supplier of intermediate inputs for the manufacturing sector. They are among the five most important professional business services measured by the amount of the intermediate inputs received by the manufacturing sector (see figure 5.1.9). Market entry regulations (e.g. requirements for profession-specific educational qualifications, membership in a chamber) and behavioural regulations (e.g. fee scales, bans on advertisement) are registered for every profession. From these sub-indicators overall indexes are derived for ev‐ ery profession and for professional business services overall. Figure 5.1.12 shows the development of the index values for Germany, Italy, France, the Netherlands and the United Kingdom. The scale reaches from no regulation (zero) to maximum regulation (six). Germany im‐ proved from 4.28 in 1998 to 2.65 in 2013. Thus it appears that in this time period massive deregulations took place. Italy and France show compara‐ ble levels of regulation. The Netherlands and the United Kingdom have OECD values slightly above and below an indicator value of one respec‐ tively, which are the lowest values in Europe. 4 V How to Assess the Economic Impact of Regulation 194 OECD Indicator on regulation in professional business services Note: All professions, including accounting, legal, architecture and engineering Source: OECD indicators of regulation in non-manufacturing sectors (NMR) A glance at the sub-indicators for the professions public and tax account‐ ing, legal, engineering and architecture show improvements for Germany in all areas in the observation period. The strongest deregulation was mea‐ sured for public and tax accountants with an improvement of the indicator value from 5.0 to 2.6. This development was mainly driven by easing the behavioural regulations (e.g. in the area of the fee scales). In the area of legal services the indicator decreased by slightly more than one grade point from 4.8 to 3.6. The underlying deregulation regarded both the be‐ havioural regulations and the market entry regulations. The indicator value for architects decreased from 4.0 to 2.8. Particularly the abolition of be‐ havioural regulations like “Interprofessionelle Zusammenarbeit”75 con‐ tributed to this. Despite the low basic level the values for engineers de‐ crease by more than one grade point from 3.3 to 1.7. This is partly due to Figure 5.1.12: 75 In Germany, “Interprofessionelle Zusammenarbeit” means the cooperation of peo‐ ple of different professions (e.g. lawyers, accountants and doctors/physicians) within one legal form of organization (“Rechtsform”). V.1 Services liberalisation in Germany 195 the abolition of requirements for “Interprofessionelle Zusammenarbeit” and lower qualification requirements. OECD Indicator on regulation for accounting, legal, ar‐ chitecture and engineering Source: OECD indicators of regulation in non-manufacturing sectors (NMR) The OECD first published the Services Trade Restrictiveness Index (STRI) in 2014, which should particularly show the international trade and investment barriers in the examined sectors. The underlying data refers to the current regulations in 34 OECD countries and six big emerging economies. The STRI values range from zero to one. Zero stands for a to‐ tally open and one for a totally closed economy. The STRI and the OECD indicators described above show certain overlaps, but they have different focuses and differ regarding their methodology. Figure 5.1.14 shows the values for Germany and other selected EU countries for engineers and ar‐ chitects as well as for lawyers and public and tax accountants (Grosso et al. 2014a and 2014b). The sub-categories are in accordance with the cat‐ egories, which are used in the international trade agreements. Figure 5.1.13: V How to Assess the Economic Impact of Regulation 196 OECD Services Trade Restrictiveness Index Note: Year 2014 Source: OECD Services Trade Restrictiveness Index Regulatory Database. As the STRI values only exist for one time point, developments cannot be shown. Nevertheless, it generally confirms the image portrayed by the OECD services regulation index. Germany, Italy and France show the highest level of regulation for all professions, whereas Germany tends to have a slightly lower level of regulation than Italy. Lower trade barriers can be found in the Netherlands and the United Kingdom. For all profes‐ sions and countries the requirements for the movement of people are the largest block in the respective regulatory index. Thereby, it has to be con‐ sidered that the OECD also captures the freedom of movement regarding non-EU states. The magnitude of economic rents in a sector of the econo‐ my can be used as an indirect indicator for regulation. The economic rent describes profits, which exceed the usual return on the used factors. Theo‐ ry suggests that these profits go back to a low competitive intensity in the market (Boone 2005). The level of economic rents can be interpreted as an Figure 5.1.14: V.1 Services liberalisation in Germany 197 indirect indicator of anti-competitive regulations. Nevertheless, these indi‐ cators should not be used as the only indicator for regulation since econo‐ mic rents can have other causes like, for example, temporary demand shifts, misuse of market power or higher shares of profit in non-incorp‐ orated firms. Net operating surpluses in selected sectors of the economy Note: Years 1995 until 2007, in per cent proportional to the production value Source: Input-output calculations of the German Federal Statistical Office, own calcu‐ lations and illustration. The net operating surplus from the input-output calculations of the Federal Statistical Office can be chosen as a rough indicator for economic rents. The net operating surplus (or the net mixed income for companies without legal personality) describes the proportion of the gross value added, which a company has at its disposal, e.g. for capital service, to pay taxes or for investments, after subtracting compensation for employees and write-offs. Between 1995 and 2007 the average net operating surplus in proportion to the production value for the whole economy was 10 per cent. In the same time period it increased from 2 to 8 per cent in the manufacturing sector, whereas it decreased from 18 to 14 per cent in the whole services sector. The average net operating surplus also decreased in the area of business services. Nevertheless, both the basis value of 34 per cent and the final value of 28 per cent were above the other aggregates. These values at least do not contradict the hypothesis that the competitive intensity in the area Figure 5.1.15: V How to Assess the Economic Impact of Regulation 198 of business services could also be lower than in the manufacturing sector, due to regulation76. In the context of the services directive there are numerous studies by the EU, which show among other things country-specific information re‐ garding the level of regulation (z. B. Monteguado 2012, European Parlia‐ mentary Research Service 2014). In addition, the European Commission recommends Germany measures to increase the competition in the area of services (EU COM 2014). Endeavours to deregulate the services market The European single market is a key element of the European integration process, whose welfare enhancing effect is commonly acknowledged. The specific design of the four fundamental freedoms (free movement of goods, people, services and capital) is being grappled with on a regular ba‐ sis. Since the early 2000s the European commission increasingly focused on the internal market for services and developed their own strategy for liberalisation. This led to the directive on services in the internal market in 2006. The European commission still sees substantial room for improve‐ ment in the services sector. This is supported by a study for the European Parliament, in which the annual costs of the non-implementation of the in‐ ternal market for services are estimated between 337 and 637 billion Euros (European Parliamentary Research Service 2014). Although such calcula‐ tion can only be used as a rough guide, there are strong theoretical argu‐ ments for further deregulation efforts (also see chapter 6). Generally the country of origin principle applies to goods if there are no common European product standards. The country of origin principle im‐ plies that goods which were manufactured in accordance with the legal standards of its country of origin can be sold in every EU member state, even if the destination country has higher product requirements. Especial‐ ly the European Court of Justice has reemphasised the importance of the country of origin principle ever since the Cassis-de-Dijon ruling on Febru‐ ary 20th 1979 (case 120-78). In the initial draft for the directive on ser‐ vices, which was developed by the former Dutch internal market commis‐ sioner Frits Bolkestein, the country of origin principle was intended as a 5 76 For further information regarding the net operating surplus, see appendix B. V.1 Services liberalisation in Germany 199 constitutive element. This led to massive protests of interest groups in the member states, which in turn led to the fact that the country of origin prin‐ ciple was formally removed from the directive. The resistance against the so called “Bolkenstein-directive” was so strong, that some partly make the draft responsible for the failure of the votes on the common European con‐ stitution in France and the Netherlands (Crespy 2010). Although the country of origin principle for services cannot be politi‐ cally implemented for now, there are currently efforts on the EU level to deregulate services and to strengthen the internal market for services. These efforts are mainly about finding common minimum quality stan‐ dards (e.g. in the form of required qualifications for the access to profes‐ sions), which will be accepted by all member states. This endeavour did not become easier since the last EU enlargement and the increased hetero‐ geneity of the member states. An important step was the passage of the directive on the recognition of professional qualifications in September 2005, which should among other purposes provide services providers with the necessary legal certainty re‐ garding the recognition of professional qualifications in EU member states. The evaluation process, which shall determine whether the goals of the directive have been achieved or whether rectifications should be im‐ plemented, presumably ends in early 2016. The EU commission therefore does a legal scrubbing with the goal to decrease regulations for the access to professions. In a first step all member states were asked to submit all regulations for the access to professions into a central database. However, not all declarations of the member states have been submitted to the com‐ mission so far. In a next step the member states should evaluate the exist‐ ing regulations for the access to professions in a peer-review-process, which should result in a national action plan for every member state. This process will be done in two waves and will presumably last until the end of 2016, whereas first intermediate results are expected in mid-2015. Moreover, the so called High-Level-Group on business services pre‐ sented its final report in April 2014. It includes numerous suggestions how the member states can strengthen business services –e.g. by further deep‐ ening the internal market for services. The High-Level-Group expects a strengthening of business services to lead to impulses for a reindustrialisa‐ tion of the EU. The High-Level-Group sees particularly great potential for increases in wealth in the increasing service orientation of the manufactur‐ ing sector, meaning the combined supply of manufactured goods and adapted services by an industrial company, and the increasing intercon‐ V How to Assess the Economic Impact of Regulation 200 nectedness of technical equipment associated with the keyword industry 4.0. A public German debate on the possibility and benefits of a continued services liberalisation is currently not existent. The few existent debates on services liberalisation in Germany are predominantly facilitated by ini‐ tiatives from Brussels e.g. in the context of the European semester. This passive attitude could not be in the best interest of German citizens, em‐ ployees and companies in the mid- and long-run as the risk exists that deregulation processes and areas will be vastly set in Brussels. As the fol‐ lowing analysis will show, it could very well be beneficial to facilitate a decentralised process from the bottom up, particularly by Germany. Economic effects of services deregulation In economic theory the decrease of regulation which limits access to a market or a profession is associated with the expectation of positive growth effects. However, it is important to notice that the theoretical mod‐ els implicitly rest upon the assumption of a functioning general legal framework. This legal framework guarantees private property, it enables the enforcement of legally valid contracts and it prevents misuse of market power. Only under these conditions can it be theoretically concluded that decisions by market actors made in a system of free competition increase the overall welfare of society. It is assumed that markets perform their al‐ location function efficiently and the state function perfectly as a regulator. The economic models are able to show general cause-and-effect rela‐ tionships. When applying the results to real life cases, one has to consider that both market processes as well as the framework by the state can differ from the theoretical ideal. Market imperfections like asymmetric informa‐ tion or externalities can legitimise state regulation. From an economic per‐ spective, it not only has to be shown convincingly that market imperfec‐ tions make interventions seem desirable, but also that the intended addi‐ tional regulation indeed leads to an increase in welfare. The following analysis will show the theoretical effects of services lib‐ eralisation and will support these results with empirical studies where ap‐ plicable. General effects of services liberalisation can be derived from this analysis. The empirical findings can be used to estimate the significance of the respective effects. Thereby the context of each study also has to be taken into consideration. 6 V.1 Services liberalisation in Germany 201 Direct effects Liberalisations in a sector initially directly influence the supply and price structure in that sector. Due to the abolition of market barriers new com‐ panies can enter the market more easily. Market entries can occur by com‐ panies operating in other markets, foreign companies or start-ups. The threat of potential competition increases the competitive pressure for in‐ cumbent companies even before potential competitors enter the market. This (potential) competition leads to the fact that companies systematical‐ ly evaluate their production process for cost saving opportunities in order to remain competitive. Apart from process innovations the probability that economies of scale (e.g. through mergers) are implemented and economies of scope are used in the production process increases with the level of competition. Economies of scope can result from the simultaneous supply of different products by using synergies during production. An example of this in the services sector could be that constraints for lawyers and tax accountants regarding “Interprofessionelle Zusammenarbeit” are decreased, which leads to the fact that legal and tax accounting services can be provided cheaper by one provider. Moreover, it is possible to use economies of scope by integrating different steps of the chain of economic gross value added. However, it is questionable whether this plays a relevant role in the services sector, as few intermediate inputs are used in the services sector. Although the usage of economies of scope plays a bigger role in the manu‐ facturing sector, in which the sale of physical goods in combination with services is becoming increasingly important in the recent past, as de‐ scribed above. Another effect of the higher level of competition due to deregulation is that companies are forced to consider consumer preferences to a greater extent. Firstly, this means a constant improvement of existing products. Secondly, this tends to lead to more innovation activities, as companies have a higher incentive to develop new products in order to be temporarily protected from competitive pressure as an entrepreneurial pioneer. It can be argued that the opening-up of the telecommunication market in Ger‐ many in the late 1990s led to a variation of existing telecommunication services (e.g. through new rate structures) and contributed to a quick ex‐ pansion of novel products (e.g. digital data transmission) for the wider population. 6.1 V How to Assess the Economic Impact of Regulation 202 The increased use of process and product innovations usually leads to decreasing prices on sales markets.77 Technically speaking the price ap‐ proaches the long term marginal costs. One has to consider that this pro‐ cess does not have a limit point in reality, as further changes are triggered by every innovation. The advantage of a higher level of competition is that suppliers have self-interest to adapt technological breakthroughs more quickly than in closed markets and thereby contribute to their diffusion. There are various pieces of evidence of the theoretically concluded pos‐ itive effects of deregulation in empirical economic research. Griffith et al. (2006) show that the establishment of the European single market led to a significant reduction of the economic rents of providers in the affected in‐ dustries and states. Aghion et al. (2004) use a simulation model on the ba‐ sis of company-specific micro data and prove that existing companies which had to face increasing competition from foreign competitors due to the EU internal market integration, had a higher productivity. They ex‐ plain this by pointing to increased efforts of established companies to dif‐ ferentiate themselves from new competitors. Increases in productivity can explain the withdrawal of less competitive companies less well. For lawyers, accountants, architects and engineers Canton et al. (2014) use econometric methods to estimate the connection between the regulatory intensity, measured by the OECD product market regulatory indicator, and the allocative efficiency as well as the profit rates in the respective sectors for the EU member states. Their results show that deregulation leads to an increase in allocative efficiency and to a decrease in profit rates in the re‐ spective sectors. Regarding the investment behaviour, a study by Alesina et al. (2005) using time series for 21 OECD countries and different regula‐ tory indexes suggests that strict regulation suppresses investments. Griffith et al. (2006) use an econometric analysis in order to examine the connection between efforts for deregulation in product markets in the context of the EU single market and the innovation activities of com‐ panies. Apart from the economic rents as an indicator for the level of com‐ petition, they used other variables to directly measure the level of compe‐ tition e.g. the information of the European commission regarding the im‐ plementation of the internal market directive in the member states. They 77 Innovations can have a price increasing component. This is the case, if innovative behaviour leads to the supply of products and services with a higher quality. Con‐ sumer can then decide, whether they are willing to pay the higher price for the higher quality. V.1 Services liberalisation in Germany 203 conclude that the innovation activities of established companies increased with stronger deregulation, whereas the competitors became less innova‐ tive. They explain this by referring to the fact that the removal of regu‐ lation generally decreases economic rents in the respective market, so that the incentive for innovation decreases for challengers. The incentives for innovation for established companies nevertheless increase, because inno‐ vations enable them to keep at least a part of their old economics rents. Aghion et al. (2006) argue that established companies are incentivised by new competitors to innovate particularly in those sectors, which are al‐ ready competitive by international comparison, as they can thereby escape competition. In contrast, established companies in internationally less competitive sectors react to new competitors with fewer innovative activi‐ ties, as they see no chance to strengthen their competitive position through innovation. It is important to note that it is a complex endeavour to empirically measure the innovative activities of companies in the services sector78. Most studies on innovation activities take the amount of patents or the ex‐ penditure for research and development as an indicator for innovation ac‐ tivities. In many services sectors these indicators play a minor role. Firstly, service processes are rarely patentable – for instance in the form of action guidelines. Secondly, many services are adapted to the wishes and prefer‐ ences of the respective customer on-sight. Patenting does not seem prof‐ itable due to the uniqueness of the innovation. For this reason many ser‐ vices providers cannot declare R&D expenditures, although they might in‐ deed spend substantial resources to develop innovations. Indirect effects Apart from the described direct effects of the reduction of regulations, more indirect effects on upstream and downstream markets are to be ex‐ pected. Companies on downstream markets profit from lower prices dur‐ ing the procurement of its primary products in liberalised markets. They can either buy the current amount for a lower price or they can buy a high‐ er amount for the same price to increase their output. If the process of 6.2 78 Maaß and Führmann (2012) describe the challenge of measuring innovations for small and medium sized companies, to which most services providers belong. V How to Assess the Economic Impact of Regulation 204 market competition functions properly, the general prices on downstream markets should decrease whereas the output increases. This channel is par‐ ticularly important for services, as they are used as intermediate input by industrial companies to a large extent (see Figure 5.1.1). On upstream markets the effects of liberalisation are not as clear. Liber‐ alisation could lead to an increase of the traded quantity and higher prices on the upstream market, if companies on the liberalised market increased their output and therefore received more intermediate inputs. This effect can at least partly be evened out by increased cost saving efforts, if this leads to more cost consciousness in the procurement of its primary prod‐ ucts. The services sector only receives a small proportion of its intermedi‐ ate inputs from other sectors while the vast majority comes from other ser‐ vices providers (see Figure 5.1.11). The effects on upstream markets are difficult to measure and the amount of literature on this is very limited. Strong empirical evidence does however exist for indirect effects on downstream markets: Forlani (2010) measures the competitive intensity of the French services sector by regarding company data. It is thereby as‐ sumed that high economic rents are an indicator for a low level of compe‐ tition. He shows that a low level of competition e.g. due to high regulation requirements has negative effects on downstream markets. Corugedo and Ruiz (2014) examine the multiplier effect of services liberalisation for France, whereas the regulatory figures are taken from Monteagudo et al. (2012). They show that liberalisations have positive effects on down‐ stream companies, which in turn positively affect other companies. More‐ over, the initiated effects in the downstream markets can in turn create more demand for services. Barone and Cingano (2011) use a cross-sectional analysis to examine the connection between the degree of regulation in the services sector, measured by the OECD services regulatory index, and the effects on downstream markets. They show that restrictive regulation in the services sector negatively affects productivity, growth of the gross value added and exports in downstream markets. The negative effect is higher the more in‐ termediate inputs come from the services sector. Based on the methodolo‐ gy of Barone and Cingano (2011), Mocci et al. (2014) show that the re‐ duction of services regulation has positive effects on downstream com‐ panies (particularly from the manufacturing sector). We could prove positive effects of services liberalisation also for Ger‐ many. The methodology is generally equal to the approach of Mocci et al. (2014) and is outlined in detail in the appendix A. Here we only outline V.1 Services liberalisation in Germany 205 the results of the fixed-effects-regression-model. The gross value added of the whole economy79 (table 5.1.1) and the manufacturing sector80 (table 5.1.2) respectively are the dependent variables. The OECD regulatory in‐ dex for “professional business services” weight with the importance of professional business services as a supplier of intermediate inputs in downstream sectors (RegS) serves as the explanatory variable.81 As con‐ trol variables we use the level of regulation in the areas of energy (RegE) and telecommunication (RegT) as well as for sector-specific employment (EMP) weighted by the importance for downstream sectors. Our calculations show that regulation of the services sector, measured by RegS, has a significant effect both on the manufacturing sector as well as the overall economy. Thereby the following relationship is valid: a de‐ creasing value for RegS leads to an increase of the gross value added in the whole economy as well as in the manufacturing sector. Since RegS is a variable composed out of two values, it is hard to interpret the value of the coefficient of RegS. Therefore, we do not go into the degree of the effect. An assessment of the effects of a change of the OECD indicators within the model will be done in chapter 8.2. As our regression model is estimat‐ ed in levels, we standardised the variables to have unit variance so that one can directly compare the estimated coefficients within the same mod‐ el, but not the coefficients of different models. 79 As the whole economy we define all three sectors of the economy (agriculture, manufacturing and services). We exclude the sectors professional services, energy and transportation for which liberalisation effects on the other sectors should be measured. 80 As the manufacturing sector we define all areas of the manufacturing sector except the construction sector as well as mining. 81 Regarding the OECD indicator see section 4. V How to Assess the Economic Impact of Regulation 206 Effects on whole economy OECD indicator & control for output, effects on value-added in whole economy Variable Coefficient Std. err. t-stat. p-value EMP 0.1255 0.0392 3.2039 0.001*** RegS -0.0691 0.0172 -4.0171 0.000*** RegE -0.0646 0.0201 -3.2067 0.001*** RegT -0.0215 0.0089 -2.4211 0.016** Output 0.4279 0.0508 8.4292 0.000*** R2 0.3685 N 49 T 9 Note: This table reports the estimation result using the OECD indicator of regulation in non-manufacturing sectors (NMR) for “Professional services” as a measure of service regulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent variable is ln VA and the sample covers the years 1998-2007. The ex‐ planatory variables are included with one time lag. Effects on manufacturing sector OECD indicator & control for output, effects on value-added in manufacturing sector Variable Coefficient Std. err. t-stat. p-value EMP 0.0991 0.0675 1.4689 0.144 RegS -0.0642 0.0196 -3.2753 0.001*** RegE -0.0531 0.0283 -1.8765 0.062* RegT -0.0166 0.0152 -1.0889 0.278 Output 0.4344 0.0704 6.1684 0.000*** R2 0.4257 N 25 T 9 Note: This table reports the estimation result using the OECD indicator of regulation in non-manufacturing sectors (NMR) for “Professional services” as a measure of service regulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent variable is ln VA and the sample covers the years 1998-2007. The ex‐ planatory variables are included with one time lag. Table 5.1.1: Table 5.1.2: V.1 Services liberalisation in Germany 207 Effects on imports and exports In most cases liberalisation measures increase competition with foreign competitors, as lower market entry barriers make it easier for them to op‐ erate in the country concerned. Generally there are more imported goods in a liberalised market. Whether domestic suppliers will be ousted depends on how successful they are in using process and product innovations to re‐ main competitive. Moreover, deregulations again have indirect effects on imports and exports of companies in downstream markets. Particularly companies, which face international competition, should profit from the resulting cost advantages. Domestic companies facing competition from foreign companies can also benefit directly from deregulation, if the existing regulation entails discriminatory provisions against domestic suppliers. A discrimination against domestic suppliers can for example result, if there are higher regu‐ lation requirements for domestic suppliers, which cannot be applied to suppliers from other EU member states due to reasons connected with EU law. A positive effect on exports can be identified if regulations are abol‐ ished, which particularly obstruct domestic suppliers when competing in‐ ternationally. Monteagudo et al. (2012) simulate the effects of the directive on ser‐ vices on imports and exports of the EU member states. They develop dif‐ ferent scenarios regarding the progress of implementation of the directive on services in the member states and consider the interdependencies of the EU economies. In a first scenario they estimate the effects of the directive on services as it is currently implemented. Thereupon imports increased to a greater extent than exports in Germany. This relation would, however, be reversed if all member states were to implement the directive on services more consequently, namely on the level of the 5 countries with the lowest intensity of regulation. The reason for this should lie in the mutual depen‐ dencies, which should lead to the same results in all member states in case of stronger and particularly comparable liberalisation. Other studies regard the effects of services liberalisation on exports in downstream markets. The already described positive effects on down‐ stream markets improve the international competitiveness of companies operating in these markets. Among others Barone and Cingano (2011) prove this for the OECD countries. Correa-López and Doménech (2014) also show positive effects of services liberalisation for large international companies in Spain. The effects on the exports of small and medium sized 6.3 V How to Assess the Economic Impact of Regulation 208 companies are also positive, but to a smaller extent. Nordas and Kim (2013) show that particularly high technology industries can profit from a liberalisation in the area of business services with regard to exports. Ac‐ cording to the findings of the authors, high quality services in the area of energy, transportation, telecommunication and finance lead to a strength‐ ening and stabilisation of manufacturing exports. Moreover, they find that the higher the stage of development of the economy is the higher are nega‐ tive effects for manufacturing exports due to disproportionate regulation in the services sector. Labour market effects Deregulations lead to a shift of economic activities. Technically speaking resources are reallocated. This induces a process of change for employees. In highly regulated sectors with a low intensity of competition, wages are generally higher than in comparable sectors without regulation (Kleiner and Krüger, 2011), because companies can achieve higher economic rents due to the lack of competitive pressure, which are part of the redistribution mass in wage negotiations. The distribution opportunities are eliminated when regulations are abolished. Wages will therefore generally adjust to wages in comparable unregulated sectors. Nevertheless, it is not implausi‐ ble that the production in the unregulated market increases, which general‐ ly leads to the creation of additional employment. Furthermore, deregula‐ tion can indirectly lead to the above described employment effects on downstream markets. Copenhagen Economics (2005) estimate in a general equilibrium model that EU-wide employment will increase by 0.3 per cent due to the imple‐ mentation of the directive on services. Badinger et al. (2008) find effects of a comparable magnitude using an econometric estimation regarding the economic effects of the directive on services. An often recited argument against deregulation refers to the loss of do‐ mestic employment due to the fact that more goods and services from for‐ eign suppliers will be demanded. This is correct if suppliers are not com‐ petitive and are not sufficiently able to strengthen their competitive pos‐ ition through process and product innovations. Under these conditions the regulation kept jobs in the affected market alive, but on grounds of higher costs for intermediate inputs the employment development in the down‐ stream markets was unfavourable compared to a situation without regu‐ 6.4 V.1 Services liberalisation in Germany 209 lation. The regulation therefore only benefited a small group, while the costs had to be paid by all other employees and consumers of the affected products. All in all it seems reasonable to assume that the abolition of regulation strengthens the international competitiveness of the whole economy and thereby also strengthens the labour market. This is also sup‐ ported by the German experience with the EU eastward expansion. Nu‐ merous studies identify mid- and long-term benefits for the whole econo‐ my after a short period of adjustment (Bug 2011, Baas et al. 2000). Impacts on consumers If deregulation leads to a higher level of competition, consumers profit as a result. These can be companies who purchase affected products as inter‐ mediate inputs or end consumers whose consumption opportunities there‐ by increase. But also products for end consumers which use intermediate inputs from a liberalised market can generally be provided for a lower price. All in all, a broad range of consumers should profit from liberalisa‐ tion activities. A common fear is that the abolition of regulation will decrease the qual‐ ity in the affected market, which would be disadvantageous for the con‐ sumer. But this is only true if the abolition of the regulation leads to a col‐ lapse of the market for high quality products. This case is known in the theoretical literature if the demand side has a significant informational dis‐ advantage (Akerlof 1970): As the consumer cannot observe the quality of the supplied good, they are only willing to pay a price according to the ex‐ pected average quality. For suppliers it is no longer profitable to supply high quality products due to the lack of willingness to pay. This lowers the expected average quality which decreases the willingness to pay even fur‐ ther. Due to this downward spiral the market for high quality products col‐ lapses. References to this scenario are often made to justify regulations in the services sector. Nevertheless, it needs to be shown that a regulation is ca‐ pable and necessary to secure a certain minimum of quality. After all there are often private market alternatives to state regulation (Shapiro 1983), since not only consumers, but also suppliers of a high quality have an interest in overcoming the information problem. Suppliers can, for exam‐ ple, voluntarily use certificates to signal certain quality standards and thereby counteract the market failure. Furthermore, liability law as an ad‐ 6.5 V How to Assess the Economic Impact of Regulation 210 ditional instrument to implement quality standards also offers opportuni‐ ties for action. In order to correspond to the primacy of the freedom to choose an occupation and the contractual freedom within the European community of law, the burden of proof for occupation-specific regulations should generally lie on the side of the regulation. This means it has to be proven that the regulation can prevent a market failure and that it cannot be assumed that less invasive private market alternatives lead to similar re‐ sults. If private alternatives to deal with information problems (and therefore for quality assurance) exist, an abolition of the regulation offers a differen‐ tiation of the quality. Regulations often have the effect of state require‐ ments for a minimum standard. These minimum standards tie suppliers and consumers. Consumers who prefer a low level of quality cannot (legally) acquire it. Only after the abolition of the regulation, suppliers can offer a broad range of different qualities, which enables them to better ful‐ fil consumer preferences. Empirical literature on the relation between (de)regulation and service quality with a focus on business services barely exists. This is due to the fact that quality is difficult to measure. Measurement problems are partic‐ ularly substantial in the services sector, as quality can usually not be ob‐ served and it is not possible to reliably infer from the visible result to the quality of services. Politico-economic considerations for services liberalisation The considerations above have shown that deregulation in the services sector can lead to positive growth effects. Of particular importance are the effects on industrial companies which are facing international competition. They could be strengthened by services liberalisation. Nevertheless, con‐ siderations for services liberalisation often face resistance. In the follow‐ ing, some possible explanations are given. Initially it can be said that the existing regulations could be welfare en‐ hancing, if they are an appropriate measure to prevent a potential market failure. In those cases it is economically reasonable to keep regulation if no equally adequate and less invasive measure exists. Nevertheless, is seems doubtful that this is accurate for all existing regulations. A first in‐ dicator of that is the fact that numerous regulations only exist in some member states (EuZFB 2014). In case of a market failure, it is reasonable 7 V.1 Services liberalisation in Germany 211 to assume that all member states would implement respective regulations. Comparisons between the regulatory regimes of the member states have to be validated further. The chambers of lawyers in Germany, for example, undertake several tasks regarding the national rules governing the profes‐ sions, which are fulfilled directly by the state in the United Kingdom. Irre‐ spective of the question on which level this task is fulfilled, the regulatory intensity of both regimes is comparable. Given an equal regulatory intensi‐ ty, the historic and cultural specifics of the member states should be taken into consideration, as an equalisation of the regulatory regimes is not a purpose in itself, but rather essential if it is needed to achieve welfare benefits. Some services professions provide services which are ascribed to con‐ tribute to general welfare to a large degree irrespective of a possible mar‐ ket failure. Lawyers contribute to the administration of justice, doctors to health care. This context has to be considered when setting the competi‐ tion framework. Nevertheless, even here it has to be proven on a case by case basis how the regulation in question serves the general welfare and why a less invasive regulatory measure does not lead to the intended out‐ come. Thereby it is hard to understand why all services of general (econo‐ mic) interest are generally exempt for EU competition law, although eco‐ nomic research considers welfare enhancing competition to be possible (Jankowski 2014, Kochskämper 2014). Even when taking these concerns into consideration it is generally recognised that numerous regulations remain which are not aimed at pre‐ venting a potential market failure or which are inadequate in preventing it. Nevertheless, resistance against efforts to deregulate even exists in these cases. Particularly in markets with a high intensity of regulation resistance is likely to be high. Here the defence of these regulations justifies special efforts by the affected actors. This reallocation of resources into lobbying can be a non-negligible part of the welfare loss, which is created by the (unjustified) regulation (Tollison 1982). Apart from the amount of the eco‐ nomic regulatory rents, the organisational capability of interest groups plays a great role in preventing deregulation activities. This is advanta‐ geous for interest groups who already have established organisations and communication structures. Examples of this are professional associations, chambers or trade unions. These groups, firstly, have the opportunity to reach and if necessary mobilise their members through existing channels and, secondly, they are integrated into many political decision making pro‐ cesses. These groups enrich the political decision making process with V How to Assess the Economic Impact of Regulation 212 their expertise regarding the conditions of the respective sector of the economy. This close involvement entails the risk that the interest groups lobby for regulation which primarily benefits its members but not the gen‐ eral public. The political dilemma is that possible negative effects in the deregulated sector can easily be attributed and that interest groups can point that out, whereas the general efficiency gains for the whole economy are unclear. The success of deregulation is therefore difficult to market po‐ litically and to convert into electoral support. Gaining electoral support be‐ comes even harder if a decrease of market barriers leads to more foreign suppliers entering the market, which would profit but are not themselves eligible to vote in domestic elections. Approaches for reform for Germany Significant efforts for liberalisation in the services sector were undertaken in the last years. Thereby substantial progress has been made. Neverthe‐ less, there is still potential for further welfare enhancing deregulation ac‐ tivities. Although the directive on the recognition of professional qualifi‐ cations has been put into place82, its implementation is in many cases not consistent considering its goal to create an internal market for services. This is true for the European level and particularly for Germany. The re‐ maining potential consists of various small regulation screws that need to be adjusted in order to make use of the full potential of the process. All of these measures together could contribute to a non-negligible ex‐ tend to the overall economic development. In many cases resistance from occupational groups is to be expected if existing regulatory rents are en‐ dangered. This applies in particular, as most measures should be small steps for which occupational groups and their representatives have an in‐ formation advantage. Overall it should, therefore, be very difficult to at‐ tract the interest of the broad public and to communicate the advantages of the sum of measures. In what follows, we explain a few principles to identify possible areas of deregulation, which firstly suggest economic benefits and which are secondly, politically realizable with limited efforts (see section 8.1). 8 82 The Federal Constitutional Court also forced liberalisations in several fundamental decisions. An example is the decision on 12.12.2006 (Az. 1BvR2576/04), which allows lawyers to take contingency fees under certain conditions. V.1 Services liberalisation in Germany 213 Thereafter, the effects of different scenarios on economic growth are de‐ scribed using an econometric estimation model (see section 8.2). In the long run a critical assessment of all specific regulation would be desirable, which evaluates both the purpose of the regulation regarding its acceptability as well as whether the chosen instrument is capable of achieving the declared goal. Moreover, it should be assessed whether al‐ ternative instruments with a lower regulatory invasiveness are suitable. This assessment has to be done separately for every sector. This cannot be carried out in this study. Approaches for reform Important impact channels for services deregulation are the associated ad‐ vantages for the manufacturing sector and especially for exporting indus‐ trial companies. Therefore, it seems worth focussing on these services sec‐ tors which are particularly often used as intermediate inputs in industrial production. Figure 5.1.9 shows which services providers deliver interme‐ diate inputs to the manufacturing sector. By far the largest source is wholesale, followed by transportation and warehousing. After that the knowledge-intense services legal services and tax accounting services as well as consultancies, under which public accountants are subsumed, as well as architecture and engineering offices follow. In order to identify those areas of business services in which welfareenhancing deregulations are potentially possible we look at the regulatory indicators used in the literature: The OECD regulatory indicators for the knowledge-intense services in the areas legal and tax accounting (including public accounting), as well as architecture and engineering professions for Germany are usually val‐ ues comparable to France and Italy. The values for the Netherlands and for the United Kingdom are, however, considerably lower (see figure 5.1.12 and figure 5.1.13). The Services Trade Restrictiveness Index (STRI) shows similar results (see figure 5.1.14). Furthermore, research by Monte‐ guardo et al. (2012) and Canaton et al. (2014) also point to further poten‐ tial for deregulation in these sectors. As argued in section 4, economic rents, for example measured by the net operational surplus, are often used as an indicator for the intensity of competition. Thereby high economic rents can be a sign for a low level of competition, which can result from anti-competitive regulation. For the 8.1 V How to Assess the Economic Impact of Regulation 214 sectors legal services and tax accounting, consulting as well as architec‐ ture and engineering offices the official statistics in Germany show a value above average compared to other knowledge-intense services (see figure 5.1.4). In combination with the above mentioned regulatory indicators this can be seen as an indicator for possible sectors with potential for deregula‐ tion. Here it has to be considered that the aim is to decrease welfare-reduc‐ ing regulation and not to decrease economic rents to a level considered “normal” for normative reasons, because the owner is also the manager and hence firm profits and labour income accrue to the same person In these professions non-incorporated firms should be the dominant form, which could partly relativize the above-average net operational surpluses. In a first assessment using statistical considerations it can be concluded that deregulation with expected positive effects on economic growth seems possible in the areas of legal and tax accounting (including public accounting) services as well as engineering and architecture offices. Be‐ fore an actual implementation of reform measures, the statistical analysis, which is based on very general assumptions, has to be complemented by a case-specific qualitative analysis. The latter takes the overall context of the regulatory measure into account.83 However, this exceeds the context of this study. After potential areas of possible deregulation were identified in a first step, a second step is to identify promising deregulation measures in those areas. The following three principles may offer some assistance: 1. Equalisation of different profession-specific regulations in comparable fields of activity on the lower level of regulation. 2. Equalisation of different state-specific regulations for single profes‐ sions on the lower level. 3. Implementation of EU guidelines on the suggested minimum level, which means no overachieving of the EU guidelines. 83 For example the indicator value for Germany in the area of legal services is 6 due to the mandated membership in a chamber, whereas the United Kingdom has the best possible indictor value of 0 due to the lack of a the mandated membership in a chamber. This does not take into consideration, that numerous tasks, which are done by the chamber of lawyers in Germany, are undertake by the state in the United Kingdom. The difference in the intensity of regulation is therefore with a huge probability overestimated by the indicator. V.1 Services liberalisation in Germany 215 Particularly the principles 1 and 2 offer the advantage that they do not de‐ mand a complete deregulation of historically grown structures, but instead do not (initially) question their basic justification. Nevertheless, they do offer potential for a reduction of the overall level of regulation. They demand equal regulatory measures or an equal regula‐ tory intensity for comparable regulatory goals. As the reference point is al‐ ways the regulation with the lower intensity of regulation used in a com‐ parable situation in Germany, fundamental concerns and cultural differ‐ ence, which might exist compared to other countries, should not matter. Principle 3 relies upon the fact that a consensus on the underlying regu‐ lation was obtained on the European level which allows foreign com‐ panies to offer their products in Germany. Fundamental concerns against this regulation (e.g. regarding quality) should not appropriate, because Germany has consented the regulation on the EU level. An assessment of the OECD regulatory indicators for professional busi‐ ness services can offer first indications, where the three principles can be applied. Both in the area of legal as well as tax and public accounting ser‐ vices the exclusive right of these professions to be allowed to carry out certain tasks, requirements for educational qualifications, a mandated membership in a chamber and limitations in “Interprofessionelle Zusam‐ menarbeit” negatively affect the indicator. In legal services this is also true for a ban on advertising and a fee scale. With architects are better indicator value are mostly prevented by requirements for educational qualifications, a mandated membership in a chamber, a fee scale and a ban on advertis‐ ing. The indicator for engineers could be improved particularly by changes in the exclusive rights of the profession, the fee scale and the ban on ad‐ vertising. Hereafter a few examples of the principles described above are out‐ lined. They are used as an illustration of the principles. However, concrete deregulation measures in these areas must be analysed on a case by case basis. The “Interprofessionelle Zusammenarbeit” is regulated differently in the respective professional laws (principle 1). Although first steps for equalisation were undertaken in the past, there are still substantial differ‐ ences, which make “Interprofessionelle Zusammenarbeit” difficult (Henssler 2009). The mentioned professions offer similar tasks, which are particularly characterised by the fact that clients must be able to trust in the confidentiality and due diligence requirements of the mandate holder. As this is a constitutive characteristic for all three professions, it is hard to V How to Assess the Economic Impact of Regulation 216 comprehend why the “Interprofessionelle Zusammenarbeit” is obstructed by diverging requirements of the respective professions. Currently the strictest professional law prevails. Therefore, “Interprofessionelle Zusam‐ menarbeit” often is no longer attractive for members of the other profes‐ sions. These limited possibilities for “Interprofessionelle Zusammenar‐ beit” complicate the offer of a comprehensive range of services from a single source in closely related areas of the economy. Possible synergistic effects remain unused. In a further step one can assess whether the cooperation between lawyers, tax and public accountants with other liberal professions like doc‐ tors or engineers in one company can also be simplified. Here the require‐ ments are very different. For example, lawyers are only allowed to work in one company with tax and public accountants. Public accountants nev‐ ertheless, can found a company with people of all occupational profes‐ sions with a right to refuse to give evidence. An explanation for these dif‐ ferent requirements is not evident, particularly since lawyers are generally allowed to work with public accountants. These restrictive requirements prevent, for example, that law firms specialise on questions in medical law and take in a doctor as a partner.84 Another example of the application of principle 1 are shareholding re‐ quirements, which differ substantially between lawyers, tax and public ac‐ countants. While external shareholding is generally prohibited for lawyers, in tax accounting mere capital holdings are permissible for members of professions suitable for association85, if they do not exceed a certain threshold. There are no explicit requirements for public accountants (Henssler 2009). The requirements of “Interprofessionelle Zusammenarbeit” also offer possible starting points for principle 3. The law about an occupational or‐ der of the chartered accountants (WPO) is one example where EU guidelines were implemented in a stricter way than required. Ac‐ cording to § 28 Abs. 1 S. 1 WPO partnerships between members of differ‐ ent professions are only accepted if the majority of the partners are public 84 The Federal Constitutional Court currently examines, whether such prohibitions are constitutional (Vorlagebeschluss des Bundesgerichtshofs vom 16.5.2013 – IIZB7/11). 85 Which professions count as suitable for association, can be seen in the respective professional law. In the case of tax accountants these are other tax accountants, public accountants and law and patent attorneys. V.1 Services liberalisation in Germany 217 accountants. Article 3 para. 4 of the directive 2006/43/EG only demands that the majority of the votes is held by public accountants. An implemen‐ tation of the directive on the minimum level would simplify “Interprofes‐ sionelle Zusammenarbeit” (Henssler 2009). Principle 1 can also be used to analyse pricing, which is designed dif‐ ferently for lawyers, public and tax accountants. Except for public accoun‐ tants the fees in the mentioned professions are regulated by a decree. But the bindingness of these fee scales varies substantially. Generally fee scales are justified by their contribution to quality assurance and by the re‐ duction of information asymmetries for consumers. If that were true, which remains to be proven, it is still unsure, why the bindingness varies between the professions. Particularly since the fee scales for architects and engineers are also justified by the same arguments and they again have different levels of bindingness. An example of principle 2, which sees a possible starting point in equalising different regulations between states, are the requirements for the coverage of professional indemnity insurance for consulting engineers, which differ considerably between states.86 In addition, there are different requirements regarding the permissibility of advertisements (Bericht der Bundesregierung zur Lage der Freien Berufe 2012). Regarding principle 2, one could also assess why doctors, dentists and psychotherapist in Bavaria are not allowed to found corporate enterprises, while this type of company is permissible for the mentioned professions in other states. Unlike in an international comparison, cultural differences and historic path dependen‐ cies cannot provide a sufficient explanation for these differences. If one does not want to shift the legislative competence away from the states, forming a government / federal state working group with the goal to equalise the requirements on a lower regulatory level could be one option. Another possible application of principle 2 can be found in the German “point of single contact” system (PSC). The PSC should provide services providers from EU member states with all the necessary information re‐ quired to be able to offer their products in Germany. Apart from the infor‐ mation function, the PSC should fulfil a mediating function and guide sup‐ pliers from other EU member states through the administration process by coordinating between the companies and the agencies in charge regarding 86 A collection of links on the relevant requirements regarding professional indemni‐ ty insurance for consulting engineers can be found here: (last accessed on January 15th 2015). V How to Assess the Economic Impact of Regulation 218 the necessary procedures. While there is a nationwide single point of con‐ tact in many EU member states, in Germany the design of the PSC lies in the competence of the states, which has led to very different ways of im‐ plementation (Icks et al. 2010). In a working paper by the EU commission (2012), the implementation of the PSC in Germany is considered to be better than average. Nevertheless, Germany is not among the best member states in any category. Moreover, it is pointed out that the quality of the PSC services varies considerably between the states. One conceivable so‐ lution would be a lowthreshold intervention in the form of a voluntary agreement between the states to create best practice guidelines. Apart from the experiences in the states one could include suggestions from EU mem‐ ber states which performed better according to the evaluation of the EU commission. Model based estimation of the welfare effects In order to estimate the effects of further services liberalisation in Ger‐ many on the net operational surplus of professional business service, we did our own econometric calculations. These are based on the model dis‐ cussed in section 6.2 and in the appendix A. Thereby we have developed three scenarios: In a conservative estimation we assume that deregulation measures are undertaken, so that the OECD regulatory indicator in the area of profes‐ sional business services decreases by half a grade point. The optimistic scenario assumes an improvement by one full grade point. Additionally, we show, which effects result from the model and if the indicator value for Germany were to drop to the level of the EU member state with the lowest values (the United Kingdom). The first two scenarios do not seem unrealistic. The third scenario will most likely not be politically feasible for the above stated reasons. Fur‐ thermore, it has to be considered that the model follows a simple linear logic, whereas a variation of the indicator value by one unit always leads to an increase of the gross value added of a certain percentage. Moreover, the forecast uncertainty of values on the edge like in scenario 3 is particu‐ larly high. The following values are therefore rough reference point and not to be considered as exact estimations. 8.2 V.1 Services liberalisation in Germany 219 In the conservative scenario the gross value added of the whole econo‐ my would increase by 0.52 per cent. For the manufacturing sector the in‐ crease would be slightly higher with 0.60 per cent. The optimistic scenario yields an increase of the gross value added in the whole economy of 1.05 per cent and of 1.21 per cent for the manufacturing sector. Assuming the OECD indicator for “professional business services” for Germany would drop to the level assumed in scenario 3, this would, under the assumptions of a linear estimation model, result in an additional gross value added of 2.02 per cent for the whole economy. For the manufacturing sector the in‐ crease would be higher with 2.33 per cent. Econometric estimation results for different deregulation sce‐ narios Effects of hypothetical reductions in regulation in the production sectors NACE M69 rev 2 “Legal and accounting activities” and NACE M71 rev 2 “Architectural and engineering ac‐ tivities” on value added, linear approximation based on the estimation results of the models A2a and A2b Corresponding OECD index value Effect on valueadded whole economy Effect on valueadded manufac‐ turing sector Actual value 2013 OECD index value professional services 2.65 Improvement of OECD index value by half a point 2.15 +0.52% +0,60% Improvement of OECD index value By one point 1.65 + 1.05% +1.21% Improvement of OECD index value level of UK 0.72 + 2.02% + 2.33% Conclusion This study has dealt specifically with the German services sector. First, we outlined the quantitative importance of the services sector in a descriptive part: in the past few decades the contribution of the services sector to the economic output increased. In 2012 one in seven Euros was generated in the services sector. Within the services sector the business services have the highest the growth dynamic. The output of the business services main‐ ly enters the further production process in the form of intermediate inputs. Table 5.1.3: 9 V How to Assess the Economic Impact of Regulation 220 Due to a differentiated statistical analysis of the division of labour be‐ tween sectors we could show that firms, which are counted as part of the manufacturing sector, produce an increasing amount of services. In 2013 on average five in ten employees in the manufacturing sector executed ser‐ vice tasks. In the field of pharmacy and chemistry even eight or rather nine in ten employees executed service tasks. This trend towards the servi‐ tisation of the manufacturing sector finds only little attention in the current debate about the supposed service gap in Germany. Because of the lack of data from other countries, we have only limited possibilities to interpret these findings. But even if we assume that the servitisation in Germany is more advanced than in other countries we can only formulate hypotheses about the causes. Cultural differences could be one reason: the German business structure is heavily influenced by owner-managed firms. We can assume that those businesses have a preference for executing important tasks in-house. A further hypothesis is that an outsourcing of services is blocked by regulations in the services sector. An evaluation of several regulatory indicators leads us to the result that significant liberalisations took place in Germany since the year 2000. They can be explained, inter alia, by the services and professional qualifi‐ cations directives, which contributed to the deepening of the common in‐ ternal market for services on the EU-level. Furthermore, within Germany decisions of the Supreme Court (Bundesverfassungsgericht) led to a re‐ duction of professional regulations. But the analysis of the available data shows that it can be assume that further deregulation potential exists with‐ in the area of the professional services (lawyers, tax and public accoun‐ tants as well as architects and engineering offices). This assumption is supported by the literature review of the evaluation of several statistical regulation measures and newer studies. We introduced three principles for the identification of promising areas for potential deregulations, which on the one hand implicitly acknowledge the historic and cultural background of the existing regulations and on the other hand still allow for liberalisations. The first principle suggests har‐ monisation of different profession specific regulations for comparable pro‐ fessions with similar regulation aims to the lowest existing level. The sec‐ ond principle aims at harmonising the profession specific regulations be‐ tween the German states to the lowest existing level. The third principle is to execute EU guidelines without going beyond the demanded regulations. An important consequence of deregulations in the services sector is the expected positive effects on downstream manufacturing markets. If the V.1 Services liberalisation in Germany 221 prices for business services drop as a result of deregulation, this will con‐ stitute a cost advantage for firms, which depend on those services as inter‐ mediate inputs. Especially firms, which compete on the international-lev‐ el, can thereby improve their competitive position. Furthermore, the con‐ sumers benefit from dropping prices and in the tendency more differentiat‐ ed product offerings. Overall an increase in economic performance can be expected. With the help of an econometric model we estimated the impact of three different deregulation scenarios on the gross value added of the overall economy and the manufacturing sector. In the course of this, we restricted our approach to deregulation of professional services. The re‐ sults show that deregulations of these services, which are heavily used by the manufacturing sector as intermediate inputs, can have significant im‐ pacts on welfare. One should notice that the econometric estimates can on‐ ly serve as a rough orientation, because of the generally associated uncer‐ tainties. Based on the approach suggested by us, policy makers can identify con‐ crete deregulation plans and evaluate their effects on welfare with the help of an individual case analysis. In the course of this, the suggested princi‐ ples can be in general used for all service areas. Due to our approach to focus on business services, we have especially chosen application exam‐ ples from the professional services. Possible further areas of application could for example be seen in services in health and social care or in the network-bound services like transport, energy and telecommunication. Be‐ fore it comes to concrete reforms one should specify and extend our pre‐ sumptions of deregulation potential, which rely on general and statistical considerations, by an individual analysis. This has to be done in order to respect the specific circumstances and to secure that liberalisation leads to an increase in welfare. Appendix Appendix A: Empirical Analysis for Germany In order to quantitatively evaluate the impact of a further liberalisation in the services sector in Germany, we follow the approach of Mocci et al. (2014), which in turn is based on the methodology of Barone and Cingano (2011). Essentially, we apply a fixed effects panel data regression in which sector specific output is regressed on a measure of services liberalisation, V How to Assess the Economic Impact of Regulation 222 some control variables and time dummies. Our regression model aims at gaining a deeper understanding of the effects that services deregulation had in Germany in the past. Based on our findings we draw some conclu‐ sions about the potential effects of a further liberalisation in the services sector. Regression model and data basis The model considers the effects of deregulation in selected services sec‐ tors on the downstream production areas which use the selected services as intermediate inputs. The dependent variable of interest is the sector spe‐ cific gross value added (VA) in real terms, computed from national inputoutput tables provided by Eurostat. The input-output calculations divide the production within an economy into different sectors and give an overview over commodity flows between the sectors.87 Our sample period is limited by the fact that the classification of the sectors as well as the cal‐ culation basis was changed in the year 2008 and that it is not possible to compare the data before and after the re-classification. Therefore, the ana‐ lysis can only be based on data until 2007. However, the last essential ef‐ forts on services deregulation in Germany took place at the beginning of the new century. Therefore, using data up to the year 2007 seems to be ap‐ propriate for analysing the systematic connection between changes of the level of regulation and output growth. In order to measure the effects of deregulation in selected services sec‐ tors, a suitable indicator is needed. Based on Barone and Cingano (2011) as well as Mocci et al. (2014) our basic model uses the OECD regulatory indexes for 1998, 2003 and 2008 as a measure for sector-specific regu‐ lation (see OECD 2013). The OECD indicator measures the extent to which policies promote or inhibit competition in the areas of non-manu‐ facturing-sectors. The indicator covers formal regulations in the areas of state control of business enterprises, legal and administrative barriers to entrepreneurship, barriers to international trade and investment. The indi‐ cator index records values on a scale from 0 to 6, where 6 represents the most restrictive-to-competition regulatory set up. The fact that before the 87 Until the change in the classification system in 2008 the production within an economy was divided according to the NACE Rev. 1.1 (2002) standard into 59 sectors. Our model uses the classification of the NACE 2002 standard. V.1 Services liberalisation in Germany 223 year 1998 no information was available on regulation in the services sec‐ tor further restricts the sample period available for the analysis. In order to obtain annual observations we linearly interpolated the indi‐ cators for the remaining years. The motivation for doing so is that the deregulation did not only take place in the years in which the correspond‐ ing variable is available, but that deregulation is a continuous process. Furthermore, it is far from obvious how long it takes after a deregulating effort has been made until the new policy goes into effect. Mocci et al. (2014) suggest that three years may be an appropriate period, but this is difficult to verify. Most likely, new policies go into effect in a continuous way, as firms adjust slowly to changing regulatory conditions. Some may anticipate these changes and react rather quickly, while others may need a few years to change their behaviour. A disadvantage of the OECD indicators is that they were only compiled for selected business services sectors. Therefore, our model can only grasp the effects of deregulation in those services for which an OECD indicator exists. These are the professional business services architecture, legal, ac‐ counting and engineering. Since our model measures the effects of deregu‐ lation on downstream sectors based on the dataset of input-output calcula‐ tions, it requires a match of the services for which an OECD indicator ex‐ ists with the corresponding productions sector in the input-output calcula‐ tions. Within the classification of the input-output tables, those profession‐ al business services belong to the production sector NACE 74 Rev. 1.1 “other business activities”. In addition to the services of architects, legal professions, accountants and engineers, the production sector NECA 74 Rev. 1.1 also includes services of management consultancy, services of ad‐ vertising and market research, employment activities (temporary work), investigation and security services as well as cleaning services. Nonethe‐ less, the professions with corresponding OECD-indicators represent a large share of the production sector NACE 74 Rev. However, it must be mentioned that the necessary assumption that the existing OECD indicators on professional business services fits for the whole production sector NACE 74 Rev. 1.1 is a drawback of the OECD indicator as a measure for the level of regulation within our model. To overcome this drawback of the OECD indicator, we propose an al‐ ternative measure for the level of regulation based on the argument that ef‐ fective deregulation should increase competitiveness in the deregulated (services) sector, which in turn should result in lower profits. Therefore, we consider the net operating surplus in the NACE 74 Rev. 1.1 sector V How to Assess the Economic Impact of Regulation 224 “other business activities”, which is again obtained from the national in‐ put-output tables, as an alternative measure of deregulation instead of the OECD indicator. In this case, our measure for the level of regulation (the net surplus) matches the entire production sector NACE 74 Rev. 1.1. Using different model variations, we analyse the effects of changes in regulation in the areas of professional business services and the two con‐ trol-groups energy88 and transportation89 on the gross value added in the following downstream production areas: – “Whole economy”: as the whole economy we define all three sectors of the economy (agriculture, manufacturing and services)90. We ex‐ clude the sectors professional services, energy and transportation for which liberalisation effects on the other sectors should be measured. – “Manufacturing sector”: as the manufacturing sector we define all ar‐ eas of the manufacturing sector except the construction sector91. Now let Xs, be the variable measuring deregulation, either the OECD indi‐ cator or the net surplus. In order to obtain a meaningful explanatory vari‐ able, Xs, is weighted by the technical coefficient wj,, which measures the importance of professional services sector s as an input for each sector j. In other words, the technical (input) coefficients wj,, represent direct back‐ ward linkages of each sector j to the professional services sector s. The technical coefficient wj,, is defined as the ratio between the costs of the inputs of service s and the value of the output of sector j. The explanatory variable RegSj, which measures the level of services regulation affecting industry sector j is computed as 88 There are OECD indicators for electricity and gas. The products of these energy services are equivalent to the production of the sector NACE 32 Rev. 1.1 in the in‐ put-output calculations. 89 There are OECD indicators for road-transport, air-transport and railway-transport. The products of these transport services are equivalent to the production of the sector NACE 60 Rev. 1.1 and NACE 62 Rev. 1.1 in the input-output calculations. 90 This excludes the sectors mining of coal and lignite (NACE 10 Rev. 1.1), which has negative gross value added, as well as mining of uranium and thorium ores (NACE 12 Rev. 1.1) and mining of metal ores (NACE 13 Rev. 1.1), in which there is no production. 91 The manufacturing sector includes the sectors NACE 10-37 Rev. 1.1 except for mining of coal and lignite (NACE 10), mining of uranium and thorium ores (NACE 12 Rev. 1.1), mining of metal ores (NACE 13 Rev. 1.1) as well as publish‐ ing, printing and reproduction of recorded media (NACE 22 Rev. 1.1). V.1 Services liberalisation in Germany 225 RegS j, t  = ∑ s wj, s, t ∙ Xs, ts .    The baseline model to be estimated then is lnVA j, t = αj + αt + β1RegS j, t − 1 + β2lnEMPj, t − 1∑k = 1 K βkZk, t − 1 + ϵ j, t,    where Zk,t for k=1,…,K are additional control variables and αj and αt rep‐ resent the fixed effects included to account for unobserved heterogeneity across sectors and time. As additional control variables we consider the weighted OECD deregulation indicators for the transportation (RegT) and the energy (RegE) sectors defined and computed in the same way as for services. The latter two variables are included as control variables in order to avoid an overestimation of the effect of deregulation in the service sec‐ tor, which is likely to be correlated with deregulation in other sectors. Sec‐ tor specific employment (EMP) is obtained from the input-output tables. Model results The estimation of our base model as described above yields the following results: Base model OECD Indicator, effects on value-added in whole economy Variable Coefficient Std. err. t-stat. p-value EMP 0.2615 0.0388 6.7339 0.000*** RegS -0.0791 0.0187 -4.2375 0.000*** RegE -0.1414 0.0195 -7.2407 0.000*** RegT -0.0242 0.0096 -2.5158 0.012** R2 0.2501 N 49 T 9 Note: This table reports the estimation result using the OECD indicator of regulation in non-manufacturing sectors (NMR) for “Professional services” as a measure of service regulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent variable is ln VA and the sample covers the years 1998-2007. The ex‐ planatory variables are included with one time lag. Table 5.1.A1a: V How to Assess the Economic Impact of Regulation 226 Base model OECD Indicator, effects on value-added in manufacturing sector Variable Coefficient Std. err. t-stat. p-value EMP 0.3054 0.0641 4.7642 0.000*** RegS -0.0673 0.0214 -3.1409 0.002*** RegE -0.1472 0.0261 -5.6413 0.000*** RegT -0.0195 0.0167 -1.1681 0.244 R2 0.3089 N 25 T 9 Note: This table reports the estimation result using the OECD indicator of regulation in non-manufacturing sectors (NMR) for “Professional services” as a measure of service regulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent variable is ln VA and the sample covers the years 1998-2007. The ex‐ planatory variables are included with one time lag. All coefficients have the expected signs and are significant. In particular, controlling for regulation levels in other sectors and for employment ef‐ fects, deregulation in the services sectors can be associated with an in‐ crease in gross value added. As our regression model is estimated in lev‐ els, we standardised the variables to have unit variance so that we can di‐ rectly compare the estimated coefficients both within and across all model variations. The interpretation of the coefficients is postponed until the end of the appendix where we predict the effect of hypothetical liberalisation measures on gross value added. Next, we consider some robustness checks of the results. In the Tables 5.1.A2a and 5.1.A2b we consider the same specification as above, but we include the lagged value of the sector specific output as an additional con‐ trol variable. This is done in order to include a variable that captures per‐ sistence in output and changing business cycle conditions. The data on sector specific output is again obtained from the input-output tables. The results shown in Table 5.1.A2a and Table 5.1.A2b confirm the findings from above. However, the estimated coefficient for the regulation in the energy sector changes quite substantially. It seems possible that the vari‐ able RegE partly captures the effect of structural change that is measured by the output variable and that the large (negative) coefficient in Table A1 is partly caused by the omission of that information. This could be ex‐ plained by the presence of an omitted variable bias in tables 5.1.A1a and Table 5.1.A1b: V.1 Services liberalisation in Germany 227 5.1.A1b. The overall fit of the regressions measured by the coefficient of determination R2 and the statistical significance of the coefficient of the output variable suggest the appropriateness of the extended model in ta‐ bles 5.1.A2a and 5.1.A2b. OECD Indicator & control for output, effects on valueadded in whole economy Variable Coefficient Std. err. t-stat. p-value EMP 0.1255 0.0392 3.2039 0.001*** RegS -0.0691 0.0172 -4.0171 0.000*** RegE -0.0646 0.0201 -3.2067 0.001*** RegT -0.0215 0.0089 -2.4211 0.016** Output 0.4279 0.0508 8.4292 0.000*** R2 0.3685 N 49 T 9 Note: This table reports the estimation result using the OECD indicator of regulation in nonmanufacturing sectors (NMR) for “Professional services” as a measure of service regulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent variable is ln VA and the sample covers the years 1998-2007. The ex‐ planatory variables are included with one time lag. OECD Indicator & control for output, effects on valueadded in manufacturing sector Variable Coefficient Std. err. t-stat. p-value EMP 0.0991 0.0675 1.4689 0.144 RegS -0.0642 0.0196 -3.2753 0.001*** RegE -0.0531 0.0283 -1.8765 0.062* RegT -0.0166 0.0152 -1.0889 0.278 Output 0.4344 0.0704 6.1684 0.000*** R2 0.4257 N 25 T 9 Note: This table reports the estimation result using the OECD indicator of regulation in non-manufacturing sectors (NMR) for “Professional services” as a measure of service regulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent variable is ln VA and the sample covers the years 1998-2007. The ex‐ planatory variables are included with one time lag. Table 5.1.A2a: Table 5.1.A2b: V How to Assess the Economic Impact of Regulation 228 A further robustness check was made concerning the choice of lag length, which was varied between 1 and 4 years. This did not qualitatively affect the results. The precise estimated results are not reported but are available from the authors upon request. As mentioned above, a potential alternative measure for services liber‐ alisation is the net operating surplus in the corresponding sectors. An addi‐ tional advantage, besides providing a robustness check of our results, is that we are not restricted in terms of data, so we can use data starting in 1995. Hence, we do not need to rely on a linear interpolation to obtain da‐ ta for the intermediate years. In tables 5.1.A3a and 5.1.A3b we considered the net–surplus in the sector “other business activities”, corresponding to the services covered by the OECD indicator. The control variables are the same as above. The results continue to hold, but the interpretation of the coefficients is difficult without further exploring possible changes our proxy for deregulation may take. To be specific, the explanatory variable RegS is a weighted average of the net surplus. In addition to that, it is un‐ clear what a typical value of this variable is and how much it varies. This also makes a comparison of the coefficients of, e.g., RegS and RegE diffi‐ cult. Furthermore, it is indistinct how much net surplus is affected by deregulation efforts in the services sector. Base model net surplus, effects on value-added in whole economy Variable Coefficient Std. err. t-stat. p-value EMP 0.3855 0.0363 10.6153 0.000*** RegS -0.0918 0.0167 -5.4824 0.000*** RegE -0.0864 0.0121 -7.1303 0.000*** RegT -0.0169 0.0067 -2.5202 0.012** R2 0.2906 N 49 T 12 Note: This table reports the estimation result using the net surplus for the sector NACE 74 Rev. 1.1 “Other business activities” as a measure of service deregulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent vari‐ able is ln VA and the sample covers the years 1995-2007. The explanatory variables are included with one time lag. Table 5.1.A3a: V.1 Services liberalisation in Germany 229 Base model net surplus, effects on value-added in manu‐ facturing sector Variable Coefficient Std. err. t-stat. p-value EMP 0.5030 0.0566 8.8929 0.000*** RegS -0.0707 0.0186 -3.8054 0.000*** RegE -0.1154 0.0185 -6.2570 0.000*** RegT -0.0395 0.0129 -3.0688 0.002*** R2 0.3630 N 25 T 12 Note: This table reports the estimation result using the net surplus for the sector NACE 74 Rev. 1.1 “Other business activities” as a measure of service deregulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent vari‐ able is ln VA and the sample covers the years 1995-2007. The explanatory variables are included with one time lag. Including output as an additional control variable again leads to an im‐ proved model fit in terms of the R2. The coefficient on RegS increases slightly, whereas again the coefficient on RegE decreases significantly (see tables 5.1.A4a and 5.1.A4b). Net surplus & control for output, effects on value-added in whole economy Variable Coefficient Std. err. t-stat. p-value EMP 0.1508 0.0336 4.4906 0.000*** RegS -0.1045 0.0139 -7.5424 0.000*** RegE -0.0356 0.0105 -3.3827 0.001*** RegT -0.0058 0.0056 -1.0446 0.297 Output 0.5695 0.0365 15.6208 0.000*** R2 0.5163 N 49 T 12 Note: This table reports the estimation result using the net surplus for the sector NACE 74 Rev. 1.1 “Other business activities” as a measure of service deregulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent vari‐ able is ln VA and the sample covers the years 1995-2007. The explanatory variables are included with one time lag. Table 5.1.A3b: Table 5.1.A4a: V How to Assess the Economic Impact of Regulation 230 Net surplus & control for output, effects on value-added in manufacturing sector Variable Coefficient Std. err. t-stat. p-value EMP 0.1784 0.0532 3.3567 0.001*** RegS -0.0804 0.0150 -5.3634 0.000*** RegE -0.0368 0.0163 -2.2622 0.025** RegT -0.0155 0.0106 -1.4673 0.144 Output 0.5674 0.0478 11.8832 0.000*** R2 0.5878 N 25 T 12 Note: This table reports the estimation result using the net surplus for the sector NACE 74 Rev. 1.1 “Other business activities” as a measure of service deregulation within the production sector NACE 74 Rev. 1.1 “Other business activities”. The dependent vari‐ able is ln VA and the sample covers the years 1995-2007. The explanatory variables are included with one time lag. In order to estimate the effects of further structural reforms in the area of professional business services a hypothetical improvement of the values of the OECD regulation indicator is assumed. On the basis of the results of our model one can conclude which effects a change in the level of regu‐ lation in the model has on the gross value added in the manufacturing sec‐ tor as well as the gross value added in the whole economy92. We consider three hypothetical scenarios. In a conservative estimation, we assume that the German level of regulation in the professional services improves by half a point. In an optimistic scenario, an improvement by one full grade point is assumed. Additionally, we consider the hypothetical case that the domestic level of regulation in the area of professional ser‐ vices reaches the average level of the leading EU member state, the United Kingdom. In 2013, the average index value for United Kingdom was 0.72. The prediction is complicated by the fact that the OECD-indicators for “professional services” do not cover the entire NACE 74 Rev. 1.1 “other business activities”. There are two possible ways to address this problem. Table 5.1.A4b: 92 The reader should be aware that the following estimates of the effect of a further deregulation have to be interpreted with care, as they rest on the assumption of lin‐ earity. Furthermore, our model does only cover the effects of services deregulation in the area of professional services, which is not whole services sector. V.1 Services liberalisation in Germany 231 For the model estimation based on the OECD indicators we had to as‐ sume that the OECD indicators have validity for the entire production sec‐ tor NACE 74 Rev. 1.1 “other business activities”. If one considers this as‐ sumption acceptable, one could determine the effect of a further deregula‐ tion in the area of professional services on gross value added based on the technical coefficients for the entire production sector NACE 74 Rev. 1.1 “other business activities”. Since the intermediate consumptions for down‐ stream sectors of the entire production sector NACE 74 Rev. 1.1 “other business activities” have a higher volume than the intermediates of the “professional services” architecture, legal, accounting and engineering (covered by the OECD indices), the resulting estimated total effects on gross value added are obviously higher (see table 5.1.A5a). Thus one im‐ plicitly assumes that the other sectors within NACE 74 Rev. 1.1 are sub‐ ject to the same deregulation as the ones covered by the OECD indices. If you follow this first approach, the value for RegS in the case of a hypo‐ thetical change of the OECD-indicator is given by RegS = ∑swj, NACE 74 rev 1,2007 ∙ Xhypothetical OECD index professional services, 2013  The assumption that the OECD indicators for “professional services” have validity for the entire production sector NACE 74 Rev. 1.1 “other business activities” was necessary because of the coarse subdivision of the inputoutput tables until the year of 2008. If such assumptions are not strictly necessary, it might be appropriate to avoid them. As the OECD-indicators are only available for the services of architects, legal professions, accoun‐ tants and engineers, it seems advisable to limit the prediction on the ef‐ fects of further deregulation to these “professional services”. In order to take account of this, we used the technical coefficients for the production sectors NACE M69 rev 2 “legal and accounting activities” and NACE M71 rev 2 “architectural and engineering activities” of the more detailed German input-output table of the year 2008 to determine the importance of deregulation in the area of these four “professional services”. Since the prediction in this case involves a smaller number of services, the resulting estimated total effects on value-added are obviously lower (see table 5.1.A5b). This results from the implicit assumption that the regulation lev‐ el in the remaining service sectors remains unchanged. If you follow our second approach, the value for RegS in the case of a hypothetical change of the OECD-indicator is given by RegS = ∑swj, NACA M69 & M71 rev 2,2008 ∙ Xhypothetical OECD index professional services, 2013  V How to Assess the Economic Impact of Regulation 232 If one assumes the hypothetical values of the OECD indicators for Ger‐ many in 2013, our model yields the following growth rates for the gross value added in the whole economy and in the manufacturing sector de‐ pending on the chosen approach: Approach one: effects of hypothetical reductions in regu‐ lation in the production sector NACE 74 Rev. 1.1 “other business activities” on value-added, linear approximation based on the estimation results of the models A2a and A2b Corresponding OECD index value Effect on valueadded whole economy Effect on valueadded manufactur‐ ing sector Actual value 2013 OECD index value professional services 2.65 Improvement of OECD index value by half a point 2.15 +1.31% +1.41% Improvement of OECD index value By one point 1.65 + 2.63% +2.84% Improvement of OECD index value level of UK 0,72 + 5.08% + 5.48% Table 5.1.A5a: V.1 Services liberalisation in Germany 233 Approach two: effects of hypothetical reductions in regu‐ lation in the production sectors NACE M69 rev 2 “legal and accounting activities” and NACE M71 Rev. 2 “archi‐ tectural and engineering activities” on value-added, lin‐ ear approximation based on the estimation results of the models A2a and A2b Corresponding OECD index value Effect on valueadded whole economy Effect on valueadded manufactur‐ ing sector Actual value 2013 OECD index value professional services 2.65 Improvement of OECD index value by half a point 2.15 +0,51% +0.60% Improvement of OECD index value By one point 1.65 + 1,01% +1.21% Improvement of OECD index value level of UK 0.72 + 2.02% + 2.33% Table 5.1.A5b: V How to Assess the Economic Impact of Regulation 234 Categorising the model results In order to make our model results comparable to Mocci et al. (2014), we estimated the model on the basis of logs and with a three year time lag. The coefficient RegS for Germany is substantially lower in our model spe‐ cifications than the values for Italy published by Mocci et al. (2014). Pro‐ jections for the effects of further deregulation in Germany, which are di‐ rectly methodologically comparable to our approach, are not be found in the literature. Most studies use general equilibrium models for projections, which cover different economic activities. These models are far more complex than our calculations. This offers the advantage to include numerous ef‐ fects into the analysis simultaneously. On the other hand the dependency of the model results on the underlying assumptions increases. Monteagudo et al. (2012) estimate the effects, which were caused by the implementa‐ tion of the services directive and the potential effects, which would result if it were implemented more consistently. Due to lacking data points after the implementation of the services directive, they also have to evade to a hypothetical scenario to calculate the effects related to the implementa‐ tion. The effect sizes of the “what if scenarios” are – if a comparison is valid at all – within the scope of our results. Varga und in ‘t Veld (2014) use the benchmark model of the European Commission in order to esti‐ mate the effects of different reforms. It is assumed that all countries imple‐ ment reforms, which put them on the level of the average of the three “best” EU countries. Due to the structure of the model, they are able to show long term changes. The effects of services liberalisation published by Varga und in ‘t Veld (2014) lie below our estimation. However in this instance a meaningful comparison of both models is once again almost im‐ possible, as – apart from the methodology – the measuring of the regula‐ tory intensity varies and baseline effects are probable. Appendix B: Operating surplus in a national and international comparison In order to categorize the net operational surpluses into selected domestic services sectors, it is useful to do a comparison with European neighbour countries. Figure 5.1.B1 compares the domestic net operational surpluses both with the EU average as well as with the surpluses of the leading neighbour countries in the OECD-NMR-ranking, namely the United King‐ V.1 Services liberalisation in Germany 235 dom and the Netherlands. Comparing the net operational surpluses of the knowledge-intense services sectors, Germany as well as the Netherlands and the United Kingdom lie roughly in accordance with the EU average of 17 per cent. Differences do however exist in the area of the professions in‐ cluded in the OECD-NMR-indicator. In the area of legal, tax accounting and consulting services the domestic net operational surpluses lie roughly on the level of the United Kingdom and the EU average, while the sur‐ pluses in the Netherlands are substantially lower. However in the area of architecture and engineering services the domestic surpluses lie consider‐ ably above those in the Netherlands and the EU average. In the area of fi‐ nancial and insurance services the domestic net operational surpluses are much lower. Net operating surplus in the area of business services in an international comparison Notes: Year 2010, in per cent proportional to production value Knowledge-intense services: media and publishing, computer and information ser‐ vices, telecommunication services, financial and insurance services, legal, tax account‐ ing and consultancy services, architecture and engineering, research and development, other liberal professions. Source: Country-specific input-output calculations, own calculations and illustration On the basis of the data of the input-output calculations a detailed break‐ down of the net operational surpluses, for example within the combined group of legal, tax accounting and consulting services, is not possible. However, a more detailed breakdown for Germany is indirectly possible for Germany through the structure survey of the services sectors of the Figure 5.1.B1: V How to Assess the Economic Impact of Regulation 236 Federal Statistical Office. On the basis of these statistics differentiated val‐ ues for the gross operational surpluses within the professions can be deter‐ mined (see figure 5.1.B2).93 In 2011 the gross operational surplus before subtracting write offs within the class of legal, tax accounting and consult‐ ing services in the area of legal services was about 49 per cent, in the area of public and tax accounting it was roughly 26 per cent, in the area of con‐ sulting services it was roughly 25 per cent and in the administration and management of companies it was roughly 8 per cent. Within the class of architecture and engineering offices the gross operational surplus in the area of architecture offices was roughly 36 per cent and in the area of en‐ gineering offices it was about 16 per cent. Gross operating surplus in the area of domestic business services Note: Year 2011, in per cent proportional to production value Source: Strukturerhebung im Dienstleistungsbereich 2011 (German services structural survey, German Federal Statistical Office), own calculations and illustration References Aghion, P.; Blundell, R.; Griffith, R.; Howitt, P.; Prantl, S. (2004): Entry and Produc‐ tivity Growth: Evidence from Microlevel Panel Data. Journal of the European Eco‐ nomic Association 2 (2-3), 265–276 Figure 5.1.B2: 93 The net operational surplus can be derived from the gross operational surplus after subtracting write offs. Regarding the input-output calculations, one can see that the write offs in the area of legal, tax accounting and consulting services are very low, as the values of the gross and the net operational surplus are almost identical. V.1 Services liberalisation in Germany 237 Aghion, P.; Blundell, R.; Griffith, R.; Howitt, P.; Prantl, S. (2009): The Effects of Entry on Incumbent Innovation and Productivity. The Review of Economics and Statistics 91 (1), 20–32 Akerlof, G.A. (1970): The Market for "Lemons": Quality Uncertainty and the Market Mechanism. In The Quarterly Journal of Economics 84 (3), 488–500 Alesina, A.; Ardagna, S.; Nicoletti, G.; Schiantarelli, F. (2005): Regulation and Invest‐ ment. Journal of the European Ecvonomic Association 3 (4), 791–825 Baas, T.; Brücker, H.; Hauptmann, A. (2009): EU-Osterweiterung: Positive Effekte durch Arbeitsmigration. Institut für Arbeitsmarkt- und Berufsforschung. IAB- Kurzbericht, 09/2009. Badinger, H.; Breuss, F.; Schuster, P.; Sellner, R. (2008): Macroeconomic Effects of the Services Directive. In Breuss,F.; Fink, G.; Griller, S. (Eds.): Services Liberalisa‐ tion in the Internal Market, vol. 6. Vienna: Springer Vienna, 125–165 Barone, G.; Cingano F. (2011): Service Regulation and Growth: Evidence From OECD Countries. The Economic Journal (121), 931–957 Bug, A. (2011): Arbeitnehmerfreizügigkeit nach der Osterweiterung der Europäischen Union. Chancen und Risiken für den Arbeitsmarkt. Deutscher Bundestag Wis‐ senschaftliche Dienste. Infobrief, WD 6 – 3010 – 001/11 Canton, E.; Ciriaci, D.; Solera, I.(2014): The Economic Impact of Professional busi‐ ness services Liberalisation. European Commission, European Economy. Economic Papers, 533. Copenhagen Economics (2005): The Economic Importance of the Country of Origin Principle in the Proposed Services Directive. Final report. Copenhagen Correa-López, M.; Doménech, R. (2014): Does anti-competitive services sector regu‐ lation harm exporters? Evidence from manufacturing firms in Spain. Edited by BB‐ VA Research. Banco Bilbao Vizcaya Argentaria. Working Paper, Nº 14/1, 14/13 Corugedo, E.F.; Ruiz, E.P. (2014): The EU Services Directive: Gains from Further Lib‐ eralization. IMF Working Paper 14/113 Crespy, A. (2010): When ‘Bolkestein’ is trapped by the French anti-liberal discourse: a discursive-institutionalist account of preference formation in the realm of European Union multi-level politics. Journal of European Public Policy 17 (8), 1253–1270 De Backer, K.; Miroudot, S.; Ragoussis, A. (2013): Manufacturing in global value chains. In: Veugelers, R. (Ed.): Manufacturing Europe’s future. Bruegel Blueprint Series, 21, 73–107 Döhrn, R. (2008): Potenziale des Dienstleistungssektors in Deutschland für Wachstum von Bruttowertschöpfung und Beschäftigung. Forschungsvorhaben des Bundesmin‐ isteriums für Wirtschaft und Technologie. Endbericht. Rheinisch-Westfälisches In‐ stitut für Wirtschaftsforschung Projektberichte, Projekt-Nr. 22/07 Eickelpasch, A. (2012): Industrienahe Dienstleistungen: Bedeutung und Entwick‐ lungspotenziale. Expertise im Auftrag der Abteilung Wirtschafts- und Sozialpolitik der Friedrich-Ebert-Stiftung Friedrich-Ebert-Stiftung (WISO Diskurs). Available online at http://library.fes.de/pdffi- les/wiso/09101.pdf V How to Assess the Economic Impact of Regulation 238 Eickelpasch, A. (2014): Funktionaler Strukturwandel in der Industrie: Bedeutung pro‐ duktionsnaher Dienste nimmt zu. DIW Wochenbericht (33), 759–770 Europäisches Zentrum für Freie Berufe (2014): Die Lage der Freien Berufe. Available online at http://www.euzfb.uni-koeln.de/fileadmin/sites/euzfb/pdf/EWSA-Studie/ complete-study-de-for- web.pdf European Commission: Detailed information on the implementation of Directive 2006/123/EC on services in the internal Market (Commission Working Staff Doku‐ ment) European Parliamentary Research Service (2014): The Cost of Non-Europe in the Sin‐ gle Market: II Single Market for Services Forlani, E. (2010): Competition in the Services sector and the Performances of Manu‐ facturing Firms: Does Liberalization Matter? CESifo Working Paper 2942 Griffith, R.; Boone, J.; Harrison, R. (2005): Measuring Competition. Advanced Insti‐ tute of Management Research. AIM Research Working Paper Series 022 Griffith, R.; Harrison, R.; Simpson, H. (2006): The link between product market re‐ form, innovation and EU macroeconomic performance. European Commission, European Economy. Economic Papers, 243. Grosso, G.M.; Lejárraga, I.; Nordås, H.K.; Gonzales, F.; Miroudot, S.; Ueno, A.; Rouzet, D. (2014): Services Trade Restrictiveness Index (STRI): Construction, Ar‐ chitecture and Engineering Services. OECD Trade Policy Papers 170. OECD Pub‐ lishing, Paris ( Grosso, G.M; Nordås, H.K.; Gonzales, F.; Lejárraga, I.; Miroudot, S.; Ueno, A.; Rouzet, D. (2014): Services Trade Restrictiveness Index (STRI): Legal and Ac‐ counting Services. OECD Trade Policy Papers 171. OECD Publishing, Paris ( Henssler, M. (2009): Die interprofessionelle Zusammenarbeit des Anwalts mit anderen Berufen. Praxisprobleme und Reformbedarf. Anwaltsblatt (10), 670–680 High-Level Group on Business Services (2014): Final Report Hill, T.P. (1977): On goods and services. Review of Income and Wealth 23 (4), 315– 338 Icks, A.; Holz, M.; Haunschild, L. (n.d.): Evaluation zum Umsetzungsstand des EA- Gesetzes in Nordrhein-Westfalen. IfM-Materialien 203 Jankowski, M. (2014): Wettbewerb und Kartellrecht in der Gesetzlichen Krankenver‐ sicherung. In Eekhoff, T.; Roth, S. J. (Eds.): Grenzgänge zwischen Wirtschaftswis‐ senschaft und Wirtschaftspolitik. Lucius & Lucius, 35–46, Stuttgart Kleiner, M.M.; Krueger, A.B. (2011): Analyzing the Extent and Influence of Occupa‐ tional Licensing on the Labor Market. IZA Discussion Paper Series 5505 Kochskämper, S. (2014): Eine Dienstleistungsrichtlinie im Gesundheitswesen Wurden die Chancen genutzt? In: Eekhoff T.; Roth, S. J. (Eds.): Grenzgänge zwischen Wirtschaftswissenschaft und Wirtschaftspolitik Lucius & Lucius, 47–56, Stuttgart Maaß, F.; Führmann, B. (2012): Innovationstätigkeit im Mittelstand – Messung und Bewertung. Institut für Mittelstandsforschung Bonn. IfM-Materialien 212 Mocci, C.; Pozzuoli, S.; Romagnoli, F.; Tinti, C. (2014): Deregulation and growth in Italy. Ministry of Economy and Finance Working Paper 3 V.1 Services liberalisation in Germany 239 Monteagudo, J.; Rutkowski, A.; Lorenzani, D. (2012): The economic impact of the Services Directive: A first assessment following implementation. European Com‐ mission, European Economy. Economic Papers 456 Nordås, H.K.; Kim, Y. (2013): The Role of Services for Competitiveness in Manufac‐ turing. OECD Trade Policy Papers 148. OECD Publishing, Paris OECD (2013): Product Market Regulation Database. Available online at www.oecd.org/econ- omy/pmr Schmidt, A.(2012): Industrie und Dienstleistungen heute: Eine Strukturanalyse der Volkswirtschaft der Bundesrepublik Deutschland. Chemie-Stiftung Sozialpartner- Akademie. Wiesbaden. Shapiro, C. (1983): Consumer Protection Policy in the United States. Zeitschrift für die gesamte Staatswissenschaft / Journal of Institutional andTheoretical Economics (139), 527–544 Tollison, R.D. (1982): Rent Seeking: A Survey. Kyklos 33 (4), 575–602 Varga, J.; in ‘t Veld, J. (2014): The potential growth impact of structural reforms in the EU. A benchmarking exercise. European Commission, European Economy. Econo‐ mic Papers 541 V How to Assess the Economic Impact of Regulation 240

Chapter Preview

References

Abstract

‘How Can We Boost Competition in the Services Sector?’ is a key question in the process of creating a more effi-cient economic environment in Germany. This book contains a collection of conference contributions on service sector reforms from members of academic institutions, ministries, the EU Commission and other organisations. The conference consisted of a keynote on the importance and implementation of structural reforms in Europe and two panels that dealt with the evaluation of past reforms in the services sector and the potential scope and effects of further reforms.

Since the 1990s, productivity growth in Germany and other Member States of the European Union has been significantly lower than in the US. The development of productivity growth in the services sector is estimated to account for two thirds of this widening gap. The European Commission advocated reforms in the services sector in its country-specific recommendations for Germany. At a conference in Berlin in July 2016, experts from various fields presented and discussed studies on service sector reforms.

With contributions by

Oliver Holtemöller, Brigitte Zypries, Joaquim Nunes de Almeida, Dirk Palige, Henrik Enderlein, Stefan Profit, Davud Rostam-Afschar, Paolo Mengano, Oliver Arentz, Erik Canton, Jochen Andritzky

Zusammenfassung

„Wie können wir den Wettbewerb im Dienstleistungssektor stärken?“ Dies ist eine Schlüsselfrage für eine größere Leistungsfähigkeit des ökonomischen Umfelds in Deutschland. Dieses Buch versammelt Konferenzbeiträge von Mitgliedern wissenschaftlicher Einrichtungen, von Ministerien, der EU-Kommission und anderen Organisationen zu Reformen im Dienstleistungssektor. Die Konferenz umfasste einen Eröffnungsvortrag zur Bedeutung und Durchführung von Strukturreformen in Europa und zwei Gesprächsforen zur Bewertung vergangener Reformen im Dienstleistungssektor und zur möglichen Reichweite sowie zu den möglichen Auswirkungen weiterer Reformen.

Die Zunahme der Produktivität ist seit den 1990er Jahren sowohl in Deutschland als auch in anderen Ländern der Europäischen Union deutlich geringer als in den USA. Es wird geschätzt, dass die Entwicklung des Produktivitätszuwachses im Dienstleistungssektor für zwei Drittel dieses zunehmenden Abstandes verantwortlich ist. Die Europäische Kommission spricht sich in ihren länderspezifischen Empfehlungen zu Deutschland für Reformen in diesem Sektor aus. Auf einer Konferenz im Juli 2016 in Berlin stellten Experten aus unterschiedlichen Bereichen Studien zu solchen Reformen vor und diskutierten deren Ergebnisse.

Mit Beiträgen von

Oliver Holtemöller, Brigitte Zypries, Joaquim Nunes de Almeida, Dirk Palige, Henrik Enderlein, Stefan Profit, Davud Rostam-Afschar, Paolo Mengano, Oliver Arentz, Erik Canton, Jochen Andritzky