ECONOMIC PERFORMANCE IN EUROPEAN ... - CiteSeerX

Economic Performance in European Telecommunications, 1978-98: A comparative study

ECONOMIC PERFORMANCE IN EUROPEAN TELECOMMUNICATIONS, 1978-98: A COMPARATIVE STUDY by Thoralf Daßler, David Parker, and David S. Saal

Thoralf Daßler, David Parker, and David S. Saal Aston Business School, Aston University, Birmingham, B4 7ET, United Kingdom

RP0108 March 2001 ISBN No: 1 85449 546 1

Aston Business School Research Institute is the administrative centre for all research activities at Aston Business School. The School comprises more than 70 academic staff organised into thematic research groups along with a Doctoral Programme of more than 50 research students. Research is carried out in all of the major areas of business studies and a number of specialist fields. For further information contact: The Director, Aston Business School Research Institute, Aston University, Birmingham B4 7ET Telephone No: (0121) 359 3611

Fax No: (0121) 333 5620

http://www.abs.aston.ac.uk/

Aston Business School Research Papers are published by the Institute to bring the results of research in progress to a wider audience and to facilitate discussion. They will normally be published in a revised form subsequently and the agreement of the authors should be obtained before referring to its contents in other published works.

0


Abstract The European telecommunications sector is undergoing major structural change in the face of new technology, privatisation and European Commission directives requiring market liberalisation. This study considers the comparative performance of the major European telecommunications operators between 1978 and 1998. This period encompasses an era of state monopolies, market liberalisation initiatives and a number of privatisations. The objectives are (1) to assess the extent so far to which market liberalisation and privatisation have impacted on the efficiency with which telecommunications services are provided in Europe; and (2) to assess changes in the performance of the different telecommunications operators over time with a view to providing an insight into the comparative efficiency performance of the different telecommunications operators in Europe. Performance is measured in terms of profit margins and labour and total factor productivity.

Key words: telecommunications, privatisation, market liberalisation, Europe, performance. JEL classification: H42; H54; L32; L33; L96; O52

1


I. Introduction Telecommunications is one of the most dynamic sectors of the European economy. In total the telecommunications industry in the EU employs over 900,000 people, has a combined turnover of US$220bn, and accounts for around 3% of Europe’s GDP. Up until the 1990s, in almost every European country competition was restricted and telecommunications were operated by state-owned utilities. The main exception was the UK, which in 1981 removed telecommunications from the control of the Post Office and established a new public corporation, British Telecommunications (BT). In November 1984 BT was privatised. At the same time, a new telecommunications operator, Mercury Communications, was licensed to compete with BT. The BT-Mercury duopoly in fixedline services continued until many of the remaining restrictions on competition were removed, after 1991 (Hunt and Lynk, 1991).1 In addition, a dedicated regulatory office, the Office of Telecommunications (OFTEL), was established to regulate telecommunication services and BT’s charges (Littlechild, 1983).

Elsewhere in Europe, telecommunications reform occurred mainly in the 1990s. The triggers for change were new technology that necessitated large-scale investment in telecommunications systems in excess of levels that some governments were willing to fund, and European Commission directives. Although the Commission stopped short of actively promoting privatisation, reflecting the EU’s official neutrality on ownership, it played an important role in advancing telecommunications market liberalisation (the development of EU policy is detailed in Schneider et al., 1994 and Curwen, 1996). In its 1987 Green Paper on telecommunications, the Commission emphasised:

‘In general, an open, competitive market for new service providers and terminal manufacturers can make a substantial contribution to the rapid spread of new services, under the current conditions of rapid development of technology and market opportunities. Given the complexity and multiplicity of the emerging telecommunications services, only the market can efficiently link the producer with the consumer. Economics knows of no other means of fulfilling this purpose 1

In more detail, the 1981 Telecommunications Act introduced competition in value added networks (VANS) and permitted companies to lease lines from BT to provide services. BT also lost its monopoly of telecom apparatus sales. From May 1987 liberalisation was extended to all value added services except telex. In July 1989 restrictions on the sale of BT private circuits were removed. In 1985 the first two cellular radio systems (‘mobile’ operators) were established and in 1987 the first cable company permitted to offer telecommunication services began operating. In 1991 the BT-Mercury duopoly of fixed line voice telecommunications was ended and later international services became fully competitive.

2


and all attempts to replace it by something else have so far failed.’ (European Commission, 1987, p.52.)

In 1988 the European Council of Ministers adopted the Commission Green Paper and confirmed the progressive introduction of a common market for telecommunications services and equipment supplies. The Green Paper was subsequently implemented through directives on terminal and network equipment liberalisation; liberalisation of services, excluding for the time being public voice telephony and operation of the basic network; separation of regulatory and operational functions; open access to networks; promotion of Europe-wide standards; and application of competition rules to the telecommunications industry. Under a July 1993 Council of Ministers resolution, full competition in voice telephony was required from 1 January 1998 for EU countries with the largest networks. Belgium and Luxembourg were permitted to delay to the year 2000 and Ireland, Spain, Portugal and Greece to 2003. However, a number of the countries chose to introduce full competition ahead of these dates.

In the face of demands for increased investment and the need for more managerial freedom so that the incumbent telecoms operators could compete with new entrants, governments in the EU decided to curtail state ownership and introduce private capital (Smith et al., 1986; Beesley & Laidlaw, 1989). Between 1985 and 1995 telecoms privatisations in Europe totalled the equivalent of US$40.5bn, putting this sector well ahead of oil and gas (US$32.7bn), banking and insurance (US$32bn) and electricity (US$26bn), in terms of the total value of asset sales (Parker, 1998, p.27). By 1998 BT in the UK and Telefónica of Spain had been wholly privatised, Deutsche Telekom, KPN of the Netherlands, Tele Danmark and Telecom Italia had been partially privatised, and planned share sales were well advanced for France Télécom, Portugal Telecom and Telia of Sweden. The sale of the first tranche of shares in Deutsche Telekom, in 1996, was the largest single privatisation issue in the EU to date. Table 1 provides details of the key dates of market liberalisation and privatisation2 (where relevant) in each of the European countries considered in this study, namely Denmark, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Spain and the UK.

2

Some enterprises had a small private shareholding before the recent privatisation programme. Also, some enterprises retain state shareholdings. The term privatisation is valid, however, because the share sale signalled the government’s intention to transfer the enterprise to the private sector. Also, in general, the enterprises were then allowed to operate as private sector companies. 3


The five largest European operators - British Telecommunications (BT), Deutsche Telekom, France Télécom, Telefónica of Spain and Telecom Italia – are therefore included in the study, along with Tele Denmark, Eircom of Ireland, OTE of Greece, the Luxembourg Enterprise des Postes and Telecommunications (henceforth, Luxembourg P&T), Koninklijke Post Telegraphy & Telecom Nederland (KPN) and Swisscom. Two non-European countries are also included in the study, the USA and Japan, for comparative purposes. In the US telecommunications operations were always privately provided, but until 1984 the sector was dominated by AT&T (‘Ma Bell’) and heavily state regulated. In 1984 AT&T was divided into eight separate regional companies (‘Baby Bells’) and a trunk and international service provider AT&T. In Japan the state-owned company Nippon Telegraph & Telephone (NTT) was privatised from 1985. In both countries, competition was gradually extended. In other words, the USA and Japan experienced

similar

technological

and

economic

pressures

to

liberalise

their

telecommunications markets as in Europe. Trends in economic performance in telecommunications in these countries should provide a useful benchmark for judging changes in performance amongst the European telecommunications operators.

Earlier studies by Foreman-Peck (1989), Foreman-Peck & Manning (1988), Duch (1991), Kwoka (1993), Parker (1994) and Martin & Parker (1997) have questioned whether privatisation leads to an immediate performance improvement in telecommunications. Kwoka, for example, finds that between 1984 and 1987 privatisation accounted for under a quarter of BT’s productivity change. More important were the growth in sales and technology. This study relates to an early period, however, and before competition developed. A recent report for the UK telecoms regulator Oftel compares the performance of BT principally with US local exchange carriers between 1995 and 1998 (Oftel, 2000). The report finds that BT operates close to the upper quartile in their sample or ninth position overall for total costs. The Oftel study, however, covers a very short period only. Also, where comparisons of unit costs are made with other European operators, the results are mixed.

Our study builds on this earlier work. It is more extensive both in terms of the period covered and the number of European telecommunications operators included. The main purposes of this study are (1) to assess the extent so far to which market liberalisation and privatisation have impacted on the efficiency with which telecommunications services are provided in Europe; and (2) to assess changes in the performance of the different telecommunications operators over time, with a view to providing an insight into the comparative efficiency of the different telecommunications operators in Europe. The

4


results should be of interest to policy makers and managers at government and company levels and to academics studying telecommunications and privatisation.

The rest of the paper is structured as follows: The following section details the profitability and productivity measures employed, as well as the data and the hypotheses we consider. Following this discussion the results are reported. Finally, the main conclusions and limitations of the study are discussed and issues for future research are outlined.

II. Performance Measures, Data, and Hypotheses The arguments for market liberalisation and privatisation have been discussed in a large number of studies and do not need to be detailed here. In essence the case for opening up monopoly state enterprises to competition draws from the economic theories of competition and property rights and the attainment of allocative and productive efficiency (Vickers & Yarrow, 1988; Parker, 2000). More specifically, competition usually benefits consumers in terms of lower prices and a higher quality of service. Competition also stimulates firms to reduce costs and innovate in terms of both new products and more efficient methods of production. Similarly, privatisation is championed as a means of achieving a more efficient allocation of resources by removing the ‘dead hand’ of state control and increasing managerial incentives to run businesses efficiently. Vickers & Yarrow (1988, p.157) suggest that the following are the main arguments for privatisation: (i) improving efficiency; (ii) reducing public sector borrowing; (iii) reducing government involvement in enterprise decision making; (iv) easing problems of public sector pay determination; (v) widening share ownership; (vi) encouraging employee share ownership; and (vii) gaining political advantage. In the EU the government debt requirements for adoption of the Euro, established at Maastricht in 1991, have been a significant driver of privatisation in recent years. But otherwise the main consideration across Europe has been improving efficiency by reducing government involvement in enterprise decision-making (ed. Parker, 1998).

The full impact of the economic reforms in the telecommunications sector will not be evident for some time. In particular, full dynamic efficiency gains are to be expected only as competition matures. Nevertheless, the industry has already undergone some major restructuring, suggesting that privatisation and market liberalisation should already have

5


had an impact on performance. Indicators of trends in efficiency over time should capture the effects on performance of the restructuring.

To assess performance, data were collected from a number of sources. The main source was the International Telecommunication Union (ITU) in Geneva, which provided data covering 1978-98 for each country on full-time staff, operating cost, depreciation, investment, traffic and number of telephone lines, as well as revenue and profit. The ITU figures are for the entire economy, but since in most of the countries a single firm accounted for all or most of the outputs and inputs, the data can be interpreted as representing wholly or largely the activities of the dominant operator.3 The ITU data were supplemented by weekly hours worked and hourly labour cost data from the Basic Statistics of the Community, published by the European statistics office Eurostat. The United Nations National Accounts: Main Aggregates and Detailed Tables were the source of investment deflators and the IMF’s International Financial Statistics provided data on interest rates and consumer prices. Where appropriate, occasional missing values were extrapolated based on existing data trends.

Finally, the Federal Communications

Commission (FCC) of the US provided further data on the US telecommunications sector.

Although the objective was to assess performance in all of the EU countries, Austria, Belgium, Finland, Portugal and Sweden were omitted from the study because of inadequate or unreliable data. Performance was measured using indicators of profitability and productivity. However, profit measures are open to serious objections especially in monopoly enterprises, because high profits reflect market dominance as well as efficient operation. Also, prices in telecommunications were set by government fiat and more recently, for example in Denmark, France, Germany, Ireland and the UK, by regulatory price caps that have an effect on realised profits. In addition, profitability is affected by state subsidies and differing accounting practices, such as including or excluding value added tax and sales tax from turnover. Nevertheless, for completeness we report below profit margin figures (profit as a percentage of revenue) in telecommunications for each of the countries studied. In judging relative performance, however, more emphasis is placed on the two sets of productivity indices – labour productivity, LP, and total factor productivity, TFP.

The calculation of both LP and TFP require the availability of a satisfactory output measure for telecommunications services. As the most consistent available data are for 3

The data were standardised to years ended 31 December where figures were reported to other accounting dates.

6


the number of main lines supplied by telecom operators, we have employed this measure. However, this measure is open to the objection that it does not necessarily reflect the quality of service (including connection times) or the volume and type of telecommunications traffic including value added services. Given this potential objection, output indices based on number of calls, call minutes, and meter pulses were also calculated as a cross-check on the results, where data permitted.4 None of these alternative measures is perfect, however. Measures of telecommunications traffic such as call minutes or number of calls do not reflect the type of call made (private, business, local, longdistance and/or international); and if meter pulses are used, it can be difficult to obtain a consistent series across countries because the billing units may vary in length.

In an early study of comparative performance in a small number of European telecommunications enterprises, undertaken in the mid-1980s, Forman-Peck & Manning (1988) used main lines as an output series alongside other indices including deflated revenues. Revenues will include services that are not reflected in other potential output indices, such as value-added and mobile services. Given this, deflated revenues were also used as a cross-check on the results for main lines. Ideally, telephone charge price deflators would be used, but these were not consistently available and therefore revenues were deflated by each country’s consumer price index. This will over or under-deflate revenues dependent upon whether telephone charges vary differently over time to general consumer prices.

Comparison of the alternative output measures, reveals that they are highly correlated (Table 2), strongly suggesting that an output index based on the number of main lines can be reliably used as the principal output series. Thus, the correlation coefficients were generally in excess of 0.9. The only exceptions were Denmark, where the correlation between calls and main lines was 0.83 and calls and deflated revenues 0.86, and the UK, where the correlation between lines and deflated revenues was 0.76. The latter result is to be expected given that since privatisation BT’s price rises have been limited to below the rate of inflation by price cap regulation. Nevertheless, alternative LP and TFP estimates employing traffic and deflated revenues did not significantly alter the conclusions drawn below, even for these two countries. We have therefore reported only results based on the more reliable main lines output measure.

4

A meter pulse is the billing unit applied by operators. 7


Labour productivity (LP) is a measure of the efficiency with which labour inputs produce outputs and is, therefore, a partial productivity measure. Labour productivity is measured as the ratio of a physical output index (Q) to a labour input index (L), with labour input adjusted for average annual hours worked in order to control for variation in the intensity of labour employment.

Total factor productivity (TFP) relates to the concept of a

production function defining the maximum output from given inputs. This is a function of engineering realities or production possibilities given existing technology, along with the management and organisation of the available resources. In this study, we follow a standard approach and calculate TFP as the ratio of Q relative to a weighted composite index of the physical inputs labour (L), capital (K), and other inputs (O), with the input weights SL, SK, and SO respectively measuring the share of each of these inputs in total economic costs.5 Further details on this method of calculating TFP are available from Heathfield & Wibe (1987), Diewert (1992), and Oum, et al. (1992).

For comparison an alternative measure of TFP was also computed by creating a composite input price deflator to deflate total costs and thereby create a proxy real input measure (used by e.g. Molyneux & Thompson, 1987, and Martin & Parker, 1997; and further described in Diewert, 1992, and Oum, et al., 1992). In this measure the weights in the composite input price index reflect the weights of each input in the imputed total costs. Hence, the imputed costs are the total of labour, other costs and the imputed rental value of capital. The TFP estimates using this method were very similar to those based on a physical input index for all of the countries except Switzerland, for which the results showed a most unlikely trend of stagnant TFP over much of the period. The results from this method of calculating TFP were therefore considered less robust and are not reported.6

Both LP and TFP will be affected by factors other than the efficiency with which enterprises are managed and this needs to be borne in mind when interpreting the results. In particular, the geography of the country (the mountainous terrain of Switzerland is a case in point), population distribution, the technology used, and legislation or trade union power that regulates manning levels, can all have an impact. Germany, for example, is known for having influential trade unions. Moreover, output and hence LP will be affected by changes in other inputs, notably capital. Total factor productivity TFP is a fuller measure that reflects the effect of labour and other inputs on output with any changes in TFP being a residual or a measure of the impact of technology and more efficient 5

The appendix sets out in detail how the inputs and the indices were computed. These results can be obtained from the authors. The TFP figures for Switzerland are also lacklustre using the favoured TFP calculation, but much less so than using the alternative measure. 6

8


management of all inputs. While TFP is superior to LP as an efficiency measure, it is more prone to error in calculation. In particular, the capital input is notoriously difficult to measure accurately.

When interpreting TFP results, it is useful to distinguish between the short-run production possibility set and the long-run set. The short-run is defined in the usual way, as a period in which one or more inputs (usually capital stock) are fixed. In the long-run, all inputs are variable and therefore optimal combinations of inputs can be selected given factor prices. In the short-run, for example, there may be excess capacity as telecommunications operators invest in anticipation of future demand, thereby deflating current productivity. The expectation is that TFP may show periods of low growth where there is high investment into capacity that is not fully utilised immediately. TFP is likely to be similarly affected by the economic cycle. Year to year TFP changes may, therefore, reflect movements along a short-run production function.

TFP measures are also affected by returns to scale. The input weights in TFP measures are ideally the proportional increase in output that a given proportional increase in inputs produces. Only with constant returns to scale will the correct elasticities correspond to the factor payment shares. Where this is not so the shares of inputs in total expenditure provide biased measures of the relevant output elasticities. Therefore, in interpreting the TFP results below, this limitation needs to be considered.7 It is to be expected that scale economies will exist in telecommunications, at least up to some output level when managerial diseconomies set in. If this is the case, the results below will be biased against finding high comparative TFP in countries with smaller telecoms operations.

In addition to competition and privatisation, technological change can also be expected to have an important impact on performance in telecommunications, in terms of generating new services (e.g. mobile communications; new value added services) and on inputs (particularly in requiring large-scale new investment in latest technology e.g. digital exchanges, satellite communications, optical fibre cable, ISDN lines etc) (Harper, 1989,

7

Other assumptions required for the method adopted to compute TFP include attainment of technical efficiency and the determination of input prices in markets that satisfy the conditions of neo-classical equilibrium. In terms of inputs, the correct weights are the proportional increase in output that a given proportional increase in inputs provides. Only under strong assumptions will expenditures correspond with these elasticities. It is worth noting, however, that averaging across years should reduce bias caused by lags in the adjustment of outputs and inputs to price changes. Also, as one purpose of the study is to identify changes in efficiency, including higher technical efficiency, the failure to satisfy the assumption of full technical efficiency is not a serious problem for our analysis. 9


p.3; Curwen, 1997, pp.26-34). Therefore, technological change can be expected to have influenced both the profitability and productivity indices over the period studied. Precise modelling of the impact of technology is not straight forward, however, and proxy indicators have to be used, such as R&D spending. In the case of telecommunications, no comprehensive data exist on R&D expenditures or other possibly proxy indicators, such as the percentage of mechanical and electronic switches, across the countries studied. Moreover, telecommunications equipment suppliers working in conjunction with telecommunication service providers mainly develop new telecommunications technology. It takes the form of innovations that can be purchased and sometimes copied by other telecommunications operators. Therefore, arguably technological change has the potential to impact on all telecommunications operators in a similar way. In other words, the extent to which operators absorb the new technologies is a function of their managerial competence and government policies when state owned. These are considerations that we wish to capture in our comparative study of efficiency. In other words, there was no need for us to model technology separately. Nevertheless, it is recognised that it would have been helpful to have said something about the precise impact of technological change on efficiency over the period, had the data existed. This is an obvious area for future research, requiring the creation of a comprehensive technology database for European telecommunications.

In most of the countries studied, market liberalisation and privatisation occurred in or from the mid-1990s and therefore at the end of the data period. Consequently, inadequate data exist to test for changes in performance over a number of years after the event. However, other studies of the impact of privatisation on performance have reported that important efficiency gains occur in the years immediately before the companies are sold off (Martin & Parker, 1997). A similar trend might be expected for market liberalisation, given that governments and companies were given notice, in the shape of EU directives, that market entry would have to be permitted. Overall, therefore, we expect the threat of competition and privatisation to have impacted on the performance of European telecommunications operators in the years running up to the reforms. We refer to this as an ‘anticipation effect’. At the same time, we would expect that in those countries that introduced reforms early, notably the UK, Denmark Japan, and the USA, performance will already have responded. In the study two main hypotheses are therefore considered:

10


(1) Market liberalisation and privatisation created an environment more conducive to management pursuing efficiency gains in telecommunications. This impacted on operators during the 1990s as management anticipated the need to improve efficiency in the face of the threat from competition and private ownership.

(2) The performance of telecommunications in those countries that reformed their telecommunications sectors first was better than in those countries that delayed reform.

III. Empirical Results Profitability Performance was first assessed by calculating changes in the profit margin in each of the countries over time. Table 3 summarises the results. In the cases of Denmark and Luxembourg reliable data do not exist prior to 1983 and hence the period covered had to be truncated.

From Table 3 it is evident that the most consistently profitable operator over the 20-year period has been Luxembourg P&T, although its profit margin shows a secular decline. Luxembourg is a small country and the telecommunications system is relatively tiny, which may explain the result. By contrast, the profitability of telecommunications in Italy has been low but steadily growing since 1991. Overall, however, three conclusions stand out. The first is the high variability in profit margins across operators and over time. There is no consistent trend of profitability across the telecommunications businesses and profitability in some of the European operations is higher than in Japan and the USA, our benchmark countries. Second, there is some evidence of a change in the trend of profitability that might be associated with market liberalisation, including anticipation effects. In particular, profitability declined slightly in the USA and Japan from the late1980s. Profit margins in the UK (i.e. BT’s profit margin) rose after privatisation but began to fall in the early 1990s, in the face of more competition and regulatory price caps.8 Third, in the countries that privatised early, namely the UK, Denmark and Japan, there is no clear sign that profit margins improved (or deteriorated) in a way significantly different to margins in countries where the main operator remained state owned. The profit margins

8

For a more detailed treatment of the impact of regulatory tightening on BT’s profits see Parker, 1997. Our results are consistent with the findings of this study. 11


in the UK, Denmark and Japan are positive after privatisation but not substantially out of line with some other countries. In a number of the countries profitability appears to have risen from the mid-1980s and declined in the 1990s, suggesting that common investment needs, government policies and market entry threats were impacting across operators irrespective of ownership. The most obvious exceptions are Greece, where market liberalisation and privatisation have been slow; Italy, where profitability is generally low, and Germany, where profit margins appear to have been under pressure from the late 1970s. In the US, where early restructuring of the always privately held Bell System occurred in 1984, the same picture is revealed. It is particularly striking that despite much hype about the benefits of privatisation in the UK, profit margins in UK telecommunications compared to that of a number of other countries studied did not noticeably improve in the first years after BT’s sell-off, in 1984.

It would be very premature, however, to draw strong conclusions on performance from this analysis of profitability. The profit margin is a crude indicator of efficiency, for the reasons outlined earlier, and will also be affected by differences between countries in accounting for profit. The results need to be considered alongside those based on our other efficiency measures, namely labour productivity and total factor productivity.

Labour Productivity Table 4 details the number of full-time staff in each of the countries’ telecommunications sectors, and average employment growth rates. Table 5 provides estimates of trends in labour productivity, after using the number of main lines as the output index, as discussed above. Part (a) provides a productivity index for each country and part (b) of the table summarises the annual percentage changes over the period. Labour productivity is reported as an index to aid comparison across countries and over time. Absolute productivity levels between countries may not be especially meaningful given geographical and traffic differences, as discussed earlier.

Most countries maintained steady upward growth in labour productivity levels, as should be expected given continued improvements in technology and the organisational structure of the industry. Moreover, this is not surprising given that labour shedding or moderate employment growth characterised almost all thirteen countries. Thus, six had negative 1979-98 average employment growth rates, while Denmark's moderate average employment growth rate of 2.2% over 1979-98 is the highest and can be largely attributed to employment growth that occurred between 1979 and 1984.

12


Despite this continued upward trend in labour productivity, the results show no consistent increase in labour productivity growth rates across all countries. Substantial differences exist between countries in their average rate of labour productivity improvement. For example, Ireland had the highest productivity growth rate across the entire 1979-98 period, averaging 9.1% per annum, although much of this growth occurred before 1995. Annual rates of growth were also high in Greece (6.2%), Spain (5.8%), the UK (5.4%), Japan (7.0%), and the Netherlands (5.0%). The lowest average growth rates were experienced in Denmark (0.9%), Switzerland (1.8%), Luxembourg (3.0%), and finally, the USA (3.0%).

Turning to sub-periods, BT’s labour productivity growth following privatisation in 1984 was initially relatively poor, at 2.4% per annum. Productivity growth greatly accelerated, however, in the first half of the 1990s, rising to almost 13% a year. This suggests that labour efficiency gains were delayed. The early 1990s saw a liberalisation of the telecommunications market that introduced considerable competition in BT’s markets. In other words, it appears to have been market liberalisation rather than privatisation per se that led to restructuring within BT and a marked improvement in labour productivity. In the case of Japan, however, there is evidence of a sustained improvement in labour productivity growth following the privatisation of NTT in 1985. So in Japan privatisation may have been more immediately beneficial. By contrast, labour productivity did not increase markedly in Switzerland in the run up to the privatisation of Swisscom in 1998, and the Netherlands, which privatised in 1994, saw a fall in labour productivity growth in both the 1990-94 and 1995-98 periods relative to 1985-89, as was also the case in Italy after privatisation in 1994. Denmark, which privatised its telecommunications enterprise in 1990, did experience an increase in average labour productivity growth from 1.5% during 1985-89 to 3.5% in 1990-94. The evidence is therefore mixed as to whether productivity growth was higher in those countries that privatised early relative to those that delayed privatisation.

Economists would usually expect productivity to improve with the arrival of competition and this appears to have been true in the UK and in Japan where annual labour productivity growth rates rose substantially after markets were opened in 1991 and 1990 respectively. However, Denmark experienced a small decline in average labour productivity growth to 1.9% during 1995-98, following its market liberalisation in 1994. Similarly, the USA experienced declining labour productivity growth after liberalisation in 1984. In the case of most of the European countries, appreciable market liberalisation occurred only towards the end of the study period (Table 1). However, there is no consistent evidence of significant performance improvements resulting from any possible

13


liberalisation anticipation effect. In the Netherlands, Ireland and France, average productivity growth rates were lower in the 1990s than in the 1980s when state monopoly provision seemed secure. In Germany, Greece, and Italy, growth rates improved in the first half of the 1990s, but remained fairly stable as the date of market liberalisation approached.

More generally, rates after 1995 are comparatively poor even though this

was the period when market liberalisation intensified in Europe. Only in Spain and Luxembourg were labour productivity growth rates markedly higher from 1995 than between 1990 and 1994. Thus, we must conclude that neither the actual introduction of market liberalisation nor its anticipation is consistently associated with improvements in labour productivity growth. However, as the introduction of market liberalisation requires the entry of new firms that must staff up in the expectation of futures sales growth, it is quite plausible that labour productivity growth will only increase after a sufficient period of time has passed to allow the potential dynamic efficiency gains of market liberalisation to be realised. Moreover, as we will see, our conclusions with regard to TFP growth further reinforce this conclusion.

Total Factor Productivity Table 6 part (a) provides TFP indices for each year since 1978 with part (b) summarising average percentage changes in sub-periods and for the entire 1979-98 period. As with the results based on labour productivity growth, while TFP levels have continued to increase in general, there is as yet no consistent evidence of substantially higher TFP growth resulting from privatisation and market liberalisation.

Thus, the six countries that

privatised and/or liberalised sufficiently early to offer adequate evidence of the postprivatisation and/or liberalisation period, provide very mixed evidence. The USA shows a progressive improvement in TFP growth rates after liberalisation in 1984.

9

The UK

provides evidence of declines in TFP growth after privatisation in 1984, but improvements after liberalisation, thereby complementing earlier studies that found little evidence of a noticeable improvement in performance in BT immediately after privatisation (ForemanPeck & Manning, 1998; Parker, 1994). In contrast, Japan provides evidence of a small increase in TFP growth rates after privatisation in 1985, but steady declines after liberalisation in 1990; while Denmark provides evidence of an increase in productivity growth after privatisation in 1990, but a small decline after liberalisation in 1994. Finally, the Netherlands, which privatised in 1994 experienced a decline in average TFP growth from 4.8% in 1990-94 to 1.8% during 1995-98, while Italy's privatisation in 1994 was followed by lower TFP growth. 9

NERA find a wide variance in cost efficiency within the US telecommunication sector in the mid-1990s (Oftel, 2000). Our results are based on aggregate figures for the US. 14


While the remaining countries in our study privatised and/or liberalised too late in order for us to consider their post-privatisation performance, we can, nonetheless, consider whether the anticipation of privatisation and liberalisation resulted in an improvement in TFP performance during 1995-98. This evidence is once again mixed, with France, Greece, and Ireland experiencing declines in TFP growth after 1995, and also revealing that TFP growth was most robust in these countries before 1990, well before privatisation and market liberalisation began.

In contrast, Germany, Luxembourg, Spain, and

Switzerland experienced increases in TFP growth during this last period, suggesting that privatisation and or the anticipation of market liberalisation was having a positive influence on TFP growth. It is, therefore, once again difficult to find a consistent pattern of performance improvement linked to either privatisation or the anticipation of market liberalisation.

It is worthwhile to note that country specific trends in TFP growth rates are sometimes contrary to trends in LP growth, as is perhaps best evidenced by Greece and Japan which generally demonstrate increasing LP growth but declining TFP growth, and the USA which demonstrates decreasing LP growth but increasing TFP growth.

We would

interpret this as suggesting that while Greece and Japan have both shed labour at a rapid rate (Table 1), these efficiencies have been more than offset by large scale capital investment projects that by 1998 had not yet resulted in commensurate increases in output. In Japan the stagnation of the economy throughout the 1990s may also help to explain the productivity results. In Greece, it is possible that the process of economic modernisation, which accelerated after joining the EU, has resulted in a level of telecommunications investment that is based on anticipated demand rather than current output levels. In contrast, our results suggest that the USA, which realised lower rates of investment growth than many of the countries in our sample, had already invested in a relatively modern telecommunications system with excess capacity by the 1980s. As a result, it was able to expand output without corresponding capital investment and rapidly increase its rate of TFP growth.

These results are therefore consistent with telecommunications companies making large capital investments in response to technological change, and the need to expand their networks in anticipation of future demand. This investment may deflate TFP growth, at least until such time as output fully responds. This in turn suggests that this study may well underestimate the longer-term productivity impacts of privatisation and liberalisation. Changes to date in inputs may be associated with anticipatory investment that will facilitate future output growth.

15

Thus, one possible interpretation of our


inconclusive findings is that a substantial period of time is required before the full dynamic efficiency gains are realised in telecommunications from Europe’s recent bout of privatisation and market liberalisation.

IV. Conclusions This paper has studied economic performance in European telecommunications over the period 1978 to 1998. Three performance measures have been used: profit margin (PM), labour productivity (LP) and total factor productivity (TFP). As the study was designed to shed light on the comparative performance of European telecommunications operators in response to the changing structure of the telecommunications industry, two main hypotheses were considered:

(1) Market liberalisation and privatisation created an environment more conducive to management pursuing efficiency gains in telecommunications. This impacted on operators during the 1990s as management anticipated the need to improve efficiency in the face of the threat from competition and private ownership.

(2) The performance of telecommunications in those countries that reformed their telecommunications sectors first was better than in those countries that delayed reform.

The results show that there is considerable variation in both the profitability and productivity of telecommunications operators across Europe, and compared with the USA and Japan – two countries introduced into the study as international benchmarks. Some countries in Europe have embraced market liberalisation and privatisation of telecommunications more enthusiastically than others. In particular, Denmark and the UK were quick to liberalise, whereas the other European countries in the study have been more hesitant and Greece has still to open its fixed-line telecoms market to competition. The form of telecommunications restructuring in member states of the European Community reflects wider national attitudes to competition and state ownership (Hulsink, 1996; Parker, 1998, p.28). Nevertheless, these different attitudes are not directly reflected in the performance differences.

The study has therefore provided scant evidence that

privatisation has had a reliable and favourable impact on economic performance in telecommunications in the absence of market liberalisation. In addition, there is no consistent evidence that market liberalisation has had a generally depressing effect on

16


profit margins or that market liberalisation, so far, has obviously led to an improvement in labour or total factor productivity growth.

The conclusion that privatisation alone does not appear to have had a marked effect on performance is consistent with certain other studies of privatised industries. These have found a complex relationship between competition, ownership and performance (e.g. Kay & Thompson, 1996; Vickers & Yarrow, 1988; Hartley, Parker & Martin, 1991; Bös, 1993; Fraquelli & Erbetta, 2000). In terms of the second hypothesis, the UK was the first European country to privatise and liberalise its telecommunications market. The UK has recorded a steady growth in TFP, but the improved performance lagged privatisation and is more closely timed to the introduction of full competition in UK telecommunications from the early 1990s.

However, as this experience has not been replicated in several

countries including Japan and Denmark which liberalised sufficiently early for firm conclusions to be drawn, this study does not provide conclusive evidence that market liberalisation is associated with productivity improvement.

The results of this study are therefore broadly consistent with earlier studies of international telecommunications, ownership and competition. Duch (1991) in a study of change in international telecommunications concluded that performance was not obviously related to ownership. Parker (1994) and Martin & Parker (1997) point to lacklustre TFP performance in BT in the years immediately after privatisation, but with performance, especially labour productivity, rising quickly in response to increased competition after 199010. In their studies profitability, expressed as rate of return on capital, was sharply higher in BT after privatisation, but fell back as competition developed. Foreman-Peck & Manning (1988, p.54), after studying the performance of six European telecoms operators in the mid-1980s (Denmark, Italy, Norway, Spain, Germany and the UK), concluded: ‘Smaller telecom organisations appear to be more productive than larger ones, and networks that have private equity capital participation do not invariably perform better than state-owned enterprises. Privatisation as such is apparently no panacea, as yet.’

At the same time, the research results must be viewed as provisional. In reviewing performance across a number of telecommunications systems research is hampered by possible inconsistencies in data, including differences in reporting standards. In addition,

10

Molyneux & Thompson (1987) report buoyant labour productivity growth in state-owned UK telecommunications between 1978 and 1985, but a poorer performance in terms of TFP. Europe Economics (1998) report labour productivity growth figures for BT since the 1980s that are higher but not dissimilar in trend to our results. 17


our measure of TFP, while a standard approach, is dependent on some restrictive assumptions, leading to possible biases, as detailed above. Equally, it is not possible to say with any certainty whether performance changes are attributable to changes in market competition or ownership over other factors, such as scale and technology. In particular, productivity in telecommunications is affected by demand and therefore levels of network loading. It is also affected by output changes. For example, in recent years mobile telephone services have expanded quickly and are not captured in our main output measure, mainlines (although they are included in an alternative output measure used, deflated revenue11). Moreover, differences in productivity growth in telecommunications may be affected by differences in national productivity, in particular the German and US economies are known to have comparatively high productivity; while Japan’s productivity growth has been slack because of persistent economic stagnation throughout much of the 1990s. These various reservations must be borne in mind when considering the research results.

In this study, we have followed a non-parametric method of TFP growth rather than a parametric approach, because our focus was primarily on identifying trends in TFP growth, and the research has therefore provided some broad indicators of performance trends across European telecommunications. However, future research would be usefully directed at studying the effect of technology and the propensity to innovate on the pace of change in European telecommunications compared to telecommunications elsewhere, especially the USA. Future parametric analysis could therefore be profitably directed at analysing the actual determinants of telecommunications TFP growth in Europe, and whether technological effort or other factors such as differences in national productivity levels have a greater impact on telecommunications productivity growth than privatisation and/or market liberalisation.

We finally suggest that our results are consistent with the argument that in a period of rapid technological change, that is triggering large-scale network investment and 11

Using deflated revenues for output, higher TFP is recorded for all countries other than France, UK, Luxembourg and Switzerland in the 1995-98 period. This suggests that using mainlines as an output measure does under-estimate recent TFP growth. However, we do not favour this measure of output because we lack reliable telecoms price deflators for each country; see the earlier discussion. Moreover, the results using deflated revenue are consistent with those using mainlines in the sense that there is no obvious, constant relationship between changes in TFP growth, privatisation or market liberalisation, except that there is stronger evidence of TFP increases after 1995. The latter finding is more commensurate with market liberalisation acting as an incentive for productivity gains, although the same caveat applies regarding the need for caution until the longerterm effects of economic restructuring result.

18


restructuring, it is to be expected that TFP growth will be depressed by capacity building. Moreover, the same argument applies where market liberalisation is leading to new market entrants and retaliatory investment by incumbent operators. This very probably explains some of the poor productivity performance in recent years. At a later date it will be possible to repeat our study using data after 1998, so that the longer-term effects of organisational changes in telecommunications can be revealed. The results of such a study are likely to be more favourable to privatisation and market liberalisation than the current study.

19


Appendix Estimation of the Capital Stock The real capital stock index (K) was estimated for each operator using the commonly applied perpetual inventory method to estimate capital stocks from data on investment: Thus it is assumed that:

I Initial Kt = I t + (1 − δ ) Kt −1 with K initial = r +δ ,

with r indicating the long-term growth rate of capital and δ

indicating the assumed

depreciation rate of capital, which were estimated for each country based on data available from the ITU database.

Real investment (I) was based on nominal investment data

provided by the ITU and subsequently deflated using the gross fixed capital deflator available from Eurostat and the National Account Statistics of the United Nations.

Determining the Input Weights used for the TFP Estimates Total economic costs (TC) were determined as:

TC = OCK + LC + COI , with COI = OpC − LC

absence of where:

OCK are the opportunity costs of capital; OpC the operating costs provided by the ITU; LC the labour costs per annum; COI are the costs of other inputs.

Other inputs encompass materials, office equipment, rental fees, maintenance and so forth.

German figures on LC per hour were affected by the reunification of the country. Before 1991 data on LC per hour were only reported for the Federal Republic, whereas afterwards figures for both parts of Germany became available. To obtain a consistent series of LC per hour, a weighted average of both countries was created.

The derivation of the OCK is based upon bond yields (i) in each country’s capital market, obtained from the IMF. However, in the case of Greece deposit rates were used in the

20


absence of data on bond yields. Until the 1990s, telecommunications companies were usually state-owned and government bond yields are a reasonable proxy for calculating the opportunity cost of capital. After privatisation, an additional equity risk premium may be relevant, yet the incumbent companies remained dominant in the market and the risk of failure to make dividend payments and repay loans was remote. The opportunity cost of capital was calculated as: OCK

t

= (it + δ )K t

Finally, the input weights were obtained as:

SK =

OCK TC

SL =

LC TC

SO =

COI TC

The Adjusted Labour Index Before 1988, hours worked per week were only available for industry as a whole and these were used on the basis that changes in hours of work in telecommunications would broadly reflect changes elsewhere in the economy. Later, Eurostat statistics report hours worked in transport and communications and these figures were used together with the figures prior to 1988 to obtain a consistent series. For Switzerland, the US and Japan, the countries’ National Statistics were used. As a cross-check, productivity calculations using numbers employed not adjusted for working hours were also made. They produced similar results to those reported in the main text.

21


References Adams, D. S. (ed., 1982) Breaking up Bell: Essays on Industrial Organisation and Regulation, New York: North-Holland. Beesley, M. and Laidlaw, B. (1989) The Future of Telecommunications, London: Institute of Economic Affairs. Bös, D. (1993) ‘Privatization in Europe: A Comparison of Approaches’, Oxford Review of Economic Policy, vol.9, no.1, pp.94-111. Bundesamt für Statistik (1982 – 2000) Statistisches Jahrbuch der Schweiz, Zürich: Verlag Neue Zuericher Zeitung. Curwen, P. (1996) ‘The Restructuring of European Telecommunications’, The Review of Policy Issues, special issue, vol.2, no.3. Curwen, P. (1997) Restructuring Telecommunications: A Study of Europe in a Global Context, Houndsmill: Macmillan. Diewert, W.E. (1992) ‘The Measurement of Productivity’, Bulletin of Economic Research, vol. 44, no. 3, pp. 163-98. Duch, R.M. (1991) Privatizing the Economy: Telecommunications Policy in a Comparative Perspective, Ann Arbor: University of Michigan Press. European Commission (1987) Towards a Dynamic European Economy: Green Paper on the Development of a Common Market for Telecommunications Services and Equipment, Com. (87) 290 final, Brussels: CEC. Europe Economics & N. Crafts (1998) Water and Sewerage Industries General Efficiency and Potential for Improvement, report for Ofwat, London: Europe Economics. Eurostat (1980 – 2000) Basic Statistics of the Community, Luxembourg: Office for Official Publications of the European Community. Federal Communications Commission (2000) Statistics of Communications Common Carriers, Washington, DC: Federal Communications Commission. Fraquelli, G. and Erbetta, F. (2000) ‘Privatization in Italy: an analysis of factor productivity and technical efficiency’, in D. Parker (ed.) Privatisation and Corporate Performance, Aldershot; Edward Elgar. Foreman-Peck, J. (1989) ‘Ownership, Competition and Productivity Growth: the Impact of Liberalisation and Privatisation on British Telecom’, Warwick Economic Research Papers 338, University of Warwick. Foreman-Peck, J. and Manning, D. (1988) “How well is BT performing? An international comparison of telecommunications total factor productivity”, Fiscal Studies, vol.9, no.3, pp.54-67. Harper, J.M. (1989) Telecommunications Policy and Management, London: Pinter.

22


Hartley, K., Parker, D. and Martin, S. (1991) ‘Organisational Status, Ownership and Productivity’, Fiscal Studies, vol.12, no.2, pp.46-60. Heathfield, D.F. and Wibe, S. (1987) An Introduction to Cost and Production Functions, London: Macmillan. Hulsink, W. (1996) ‘Persistent Divergence in a Converging Industry: A Comparison of Telecommunications Restructuring in France, the Netherlands and the United Kingdom’, mimeo, Science Policy Research Unit, University of Sussex. Hunt, L. and Lynk, E. (1991) ‘Competition in UK Telecommunications: Restructure BT?’, Fiscal Studies, vol.12, no.3, pp.73-87. ITU (1999) ITU World Telecommunication Indicators ’98 (5th ed.), Geneva: ITU. IMF (1999) International Financial Statistics, Washington, DC: International Monetary Fund. Kay, J. and Thompson, D.J. (1986) ‘Privatisation: a Policy in Search of a Rationale’, Economic Journal, vol.96, March, pp.18-32. Kwoka, J. (1993) ‘The effects of divestiture, privatisation and competition in US and UK telecommunications’, Review of Industrial Organisation, vol.8, no.1, pp.49-61. Littlechild, S.C. (1983) Regulation of British Telecommunications’ Profitability, London: HMSO. Martin, S. and Parker, D. (1997) The Impact of Privatisation: Ownership and corporate Performance in the UK , London: Routledge. Molyneux, R. and Thompson, D. (1987) ‘Nationalised Industry Performance: still third rate?’, Fiscal Studies, vol.8, no.1, pp.48-82. Oftel (2000) The comparative efficiency of BT: A report for Oftel prepared by NERA, London: Office of Telecommunications. Oum, T.H., Tretheway, M.W., and Waters II, W.G. (1992) ‘Concepts, Methods, and Purposes of Productivity Measurement in Transportation’, Transportation Research – A, vol. 26A, no. 6, pp. 493-505. Parker, D. (1994) ‘A Decade of Privatisation: the Effect of Ownership Change and Competition on British Telecom’, British Review of Economic Issues, vol.16, no.40, pp.87-113. Parker, D. (1997) ‘Price cap regulation, profitability and returns to investors in the UK regulated industries’, Utilities Policy, vol.6, no.4, pp.303-315. Parker, D. (1998) ‘Privatisation in the European Union: an overview’, in D. Parker (ed.) Privatisation in the European Union: Theory and Policy Perspectives, London: Routledge. Parker, D. (2000) ‘Introduction’, in D. Parker (ed.) Privatisation and Corporate Performance, Aldershot; Edward Elgar.

23


Privatisation International (1997) Utility Regulation 1997, London: Privatisation International & Centre for the study of Regulated Industries. Privatisation International (2000) Utility Regulation in the EU, London: Privatisation International & Centre for the study of Regulated Industries Schneider, V., Dang-Nguyen, G. and Werle, R. (1994) ‘Corporate Actor Networks in European Policy-making: Harmonizing Telecommunications Policy’, Journal of Common Market Studies, vol.32, no.4, pp.473-98. Smith, P.E., Barnard, J.M. and Smith, G. (1986) ‘Privatisation and Cultural Change: a case study of management development in British Telecom’, Journal of Management Development, vol.5, no.2, pp.51-61. Statistics Bureau / Management and Coordination Agency (1980 – 1999) Japan Statistical Yearbook, Tokyo: Statistics Bureau / Management and Coordination Agency. United Nations (1982 – 1999) National Accounts Statistics: Main Aggregates and Detailed Tables, New York: United Nations. US Bureau of Census (180 – 1998) Statistical Abstract of the United States, Washington DC: US Bureau of Census. Vickers, J. and Yarrow, G. (1988) Privatization: An Economic Analysis, Cambridge, Mass.: MIT Press.

24


Table 1: Privatisation and Market Liberalisation in Telecommunications: Key Dates

Country: Dominant Operator

Year of Privatisation (1)

Year of Market Liberalisation (2)

Denmark: Tele Danmark (3)

1990

1994

France: France Télécom

1997

1998

Germany: Deutsche Telekom

1996

1998

Greece: OTE

1996

2001

Ireland: Eircom

1999

1998

Italy: Telecom Italia (4)

1994

1998

Japan: NTT

1985

1990

Luxembourg: P&T

No privatisation to date

1997

Netherlands: KPN

1994

1997

Spain: Telefónica

1995

1998

Switzerland: Swisscom

1995

1998

UK: BT

1984

1991

Not applicable

1984

USA: AT&T and ‘Baby Bells’

(1) Date of privatisation is the year of the first major share sale to the private sector. Note that some enterprises had a small private shareholding before the recent privatisation programmes. This is ignored because it did not affect the status of the enterprise as state owned and controlled. Also, some enterprises retain state shareholdings; for example, the Japanese Ministry of Finance still holds a 53% stake in NTT. Privatisation is, therefore, defined as the first significant share sale that signalled transfer to the private sector. Generally the enterprises were then allowed to operate as private sector companies. (2) The date of market liberalisation is the year when the first significant competition was allowed. In a number of the countries, e.g. UK, Netherlands and Spain, competition was phased in from an earlier date – in the UK from 1984 and the Netherlands and Spain from the late 1980s. Significant competition in fixed line voice telephony occurred from the dates listed, but it is accepted that putting a precise date on market liberalisation can be problematic. (3) Prior 1990 the Danish market consisted of four regional telecommunications operators, with mixed public-private ownership, and one cross-regional and international operator, part of the Danish post office. These operators were merged to form Tele Danmark in 1990, with 49 percent of its shares in private hands. (4) In 1994, Telecom Italia was formed by merging five different carriers in which the state had been a major shareholder. Source: Privatisation International (1997 & 2000).

25


Table 2: Correlation of Alternative Output Indices: 1978-98

Denmark France Germany Greece Ireland Italy Japan Luxembourg Netherlands Spain Switzerland UK (2) USA

Traffic/Main Lines (1) 0.83 0.98 0.98 0.99 0.99 0.96 0.96 0.98 0.96 0.93 0.98

Traffic/Deflated Revenues 0.86 0.90 0.94 0.96 0.92 0.96 0.91 0.95 0.99 0.91 0.91

0.99

0.99

Main Lines/Deflated Revenues 0.93 0.94 0.96 0.93 0.96 0.95 0.91 0.98 0.96 0.99 0.97 0.76 0.97

(1) Traffic refers to calls, minutes of calls or meter pulses, depending on the data availability for the economy. (2) No traffic index was available for the UK.

Table 3 Profit Margins in Telecommunications: 1978-98 (Percentage) 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

DK France 18.56 17.64 17.82 3.71 4.37 -1.42 9.57 8.64 13.42 13.71 11.42 7.88 6.50 9.71 5.34 2.04 8.06 4.81 9.92 5.33 4.18 1.76 5.80 2.67 9.49 3.74 36.94 2.74 17.14 1.88 14.03 1.39 7.43 9.49 7.60 9.33

Ger 25.82 17.60 1.25 9.73 10.34 10.34 10.36 10.60 9.79 7.71 6.24 5.11 2.82 4.70 3.00 -2.94 2.02 7.55 2.35 4.13 5.03

Greece Ireland 9.27 -20.36 4.44 -14.26 -3.03 -10.21 -17.16 -6.35 7.82 -5.14 10.86 -18.53 5.95 -11.47 1.59 -3.53 16.90 2.33 16.37 8.51 21.27 10.21 18.51 10.96 18.12 10.46 18.93 6.26 14.09 7.62 16.08 4.03 15.90 5.95 18.85 7.94 19.62 9.47 19.89 11.06 15.87 6.32

Italy 8.05 -8.28 -5.36 5.76 2.20 5.15 3.48 0.43 5.95 5.61 4.02 4.20 n/a 4.25 3.84 3.82 4.96 7.21 8.63 8.94 11.62

Japan 11.86 10.55 9.15 8.58 8.75 7.77 12.69 13.87 14.46 16.79 12.71 10.10 9.22 11.00 7.90 4.28 1.52 2.05 2.40 2.24 4.83

Lux

60.49 53.37 42.46 41.42 44.62 56.33 52.99 49.43 57.52 20.29 29.07 26.29 22.68 29.88 36.25

Neth Spain Switz UK USA 10.73 1.30 18.73 5.10 11.65 10.32 1.49 18.12 3.91 12.08 5.85 1.39 17.92 7.17 12.36 5.31 1.08 15.98 6.36 10.98 6.65 0.91 13.03 12.12 8.74 8.42 0.70 13.48 12.97 8.88 11.02 0.62 16.49 12.78 15.06 13.10 0.59 16.78 13.75 13.45 16.39 0.71 16.38 15.17 12.37 17.13 0.81 14.91 15.55 12.17 15.06 0.86 13.49 13.19 11.00 27.05 0.84 11.43 15.07 9.41 15.16 0.98 11.46 15.72 8.98 12.79 0.96 7.06 14.41 8.64 12.02 0.98 9.03 9.53 7.77 11.59 0.79 8.16 8.76 6.85 10.95 0.75 10.89 9.61 10.90 11.62 0.80 13.60 10.63 7.88 12.30 0.79 13.71 10.70 10.23 9.43 0.68 -3.09 10.73 9.03 9.44 0.67 19.96 15.15 9.90

n/a = not available; DK=Denmark, Ger=Germany, Lux=Luxembourg, Neth= Netherlands, Switz=Switzerland.

26


Table 4: Employment in Telecommunications: 1978-98 (a) Number of Employees 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

DK France 11,950 150,000 12,396 158,000 12,665 156,400 12,734 159,800 12,685 163,900 12,668 165,300 17,024 166,200 17,024 166,788 18,492 165,198 18,997 163,682 18,556 159,521 18,110 157,313 17,710 155,814 17,849 156,110 17,195 155,300 16,355 154,900 16,280 152,455 16,081 151,448 16,126 164,720 17,170 165,042 17,336 169,099

Ger 183,000 190,000 195,000 199,774 203,751 204,718 207,693 212,364 214,349 216,020 216,156 216,210 212,205 230,000 232,600 233,600 227,600 229,700 222,000 219,200 219,700

Greece Ireland Italy Japan 28,700 17,698 101,000 334,750 28,800 18,440 102,000 334,015 30,200 18,626 104,000 333,887 31,680 18,465 104,000 330,614 31,148 17,860 104,480 325,825 30,995 16,797 107,040 321,825 30,602 15,761 108,782 313,712 30,571 14,968 109,792 306,637 29,595 14,463 110,232 301,269 29,444 14,260 109,680 289,500 30,327 13,841 113,676 279,244 29,654 13,519 116,391 273,438 28,086 13,212 117,986 267,603 27,593 13,078 120,300 256,241 26,716 12,838 120,000 243,507 26,349 12,150 111,500 225,603 26,140 11,684 105,500 215,360 24,581 11,251 101,900 217,684 23,808 11,628 100,005 191,636 22,741 12,050 94,101 174,625 21,925 12,156 93,391 236,200

Lux 621 631 629 634 639 651 569 675 745 750 775 780 812 845 850 860 870 800 816 828 861

Neth Spain Switz UK USA 26,766 68,620 15,969 237,550 994,800 27,395 68,840 16,140 245,050 1,048,200 27,902 70,610 16,323 246,087 1,072,200 27,936 70,960 16,781 245,953 1,077,300 27,457 71,170 17,292 246,519 1,071,800 27,677 71,461 17,656 243,037 1,010,300 27,760 71,199 18,019 236,838 953,400 28,774 72,086 18,326 234,595 920,700 29,674 63,021 18,733 236,500 883,400 29,833 63,311 18,159 242,600 901,900 29,142 66,062 18,920 245,375 901,100 28,977 71,155 19,715 231,600 885,900 29,262 75,350 20,170 220,975 913,000 30,819 75,499 20,705 191,250 909,200 32,327 74,437 20,857 173,000 885,200 31,981 74,340 20,521 157,250 879,000 31,671 72,207 19,543 147,000 893,400 30,548 69,570 20,219 142,750 899,700 29,690 67,216 21,951 147,250 897,700 32,088 64,109 22,170 153,500 921,700 32,547 61,107 21,946 178,250 1,021,800

(b) Average Annual Percentage Change

79-84 85-89 90-94 95-98 79-98

DK France 6.72 1.74 1.33 -1.09 -2.09 -0.62 1.62 2.69 2.15 0.63

Ger 2.14 0.81 1.11 -0.87 0.95

Greece Ireland 1.11 -1.84 -0.60 -3.01 -2.48 -2.86 -4.29 1.04 -1.30 -1.81

Italy 1.25 1.37 -1.88 -2.98 -0.35

27

Japan -1.07 -2.71 -4.65 3.87 -1.39

Lux -1.31 6.73 2.22 -0.15 1.82

Neth 0.62 0.89 1.83 0.79 1.02

Spain 0.62 0.24 0.34 -4.09 -0.49

Switz 2.04 1.85 -0.14 3.00 1.64

UK -0.03 -0.41 -8.64 5.16 -1.24

USA -0.62 -1.43 0.19 3.50 0.20


Table 5: Labour Productivity: 1978-98

(a) Index of labour productivity (1995=100) 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

DK France Ger Greece Ireland Italy 93.12 35.07 46.91 35.95 19.39 46.76 93.98 38.89 49.28 37.98 20.34 48.20 94.84 44.96 52.23 37.97 23.05 51.20 96.83 49.47 54.93 38.13 25.20 56.39 99.70 54.42 56.58 40.76 27.62 59.21 101.87 58.61 58.92 43.69 31.81 63.34 77.65 61.80 60.05 47.51 36.19 64.95 79.95 64.21 61.52 50.44 40.84 68.86 75.94 67.48 62.55 54.80 44.68 70.72 76.14 70.83 64.56 57.75 48.21 73.12 80.12 75.88 67.05 58.53 53.46 72.96 83.29 80.47 70.92 62.67 58.70 74.18 86.83 84.04 81.33 68.97 64.92 77.10 86.85 86.80 77.17 73.72 70.07 76.13 92.06 90.13 82.42 80.87 75.40 81.24 98.82 93.25 85.95 85.92 84.48 89.12 98.91 96.94 92.87 90.03 92.20 95.41 100.00 100.00 100.00 100.00 100.00 100.00 106.78 93.36 107.51 106.33 103.50 103.59 103.05 94.95 112.48 112.94 102.73 111.15 106.51 93.74 114.25 119.67 109.06 114.12

Japan 36.08 37.22 37.89 39.36 41.19 42.79 45.14 47.61 50.12 54.16 58.47 62.78 67.47 73.62 81.07 91.75 98.41 100.00 114.40 127.15 137.64

Lux 65.68 67.12 69.12 70.66 72.18 72.63 72.54 74.31 69.64 70.67 70.19 76.36 76.40 76.63 82.86 86.31 88.05 100.00 108.52 115.83 117.61

Neth 39.37 41.75 44.46 47.25 50.36 52.68 55.59 56.43 57.37 59.72 65.58 81.21 86.41 85.13 84.87 88.40 92.76 100.00 103.85 97.73 100.70

Spain 40.64 45.06 47.89 51.22 53.50 57.19 61.74 64.81 77.46 78.98 79.47 77.75 76.89 80.77 85.18 88.58 93.50 100.00 105.42 113.70 123.15

Switz 74.53 76.50 78.55 79.34 79.86 81.53 82.47 83.74 85.72 91.94 91.60 92.01 93.48 94.69 96.40 99.88 97.19 100.00 100.05 101.60 105.15

UK 36.18 36.77 38.38 40.20 41.43 43.55 45.87 47.45 48.90 49.88 51.29 51.68 55.72 66.16 75.61 84.79 93.86 100.00 101.29 100.59 101.73

USA 55.10 54.12 55.28 61.77 63.97 69.42 74.43 80.10 85.39 86.82 90.20 85.12 86.16 88.61 93.83 95.14 96.01 100.00 103.03 104.59 98.31

(b) Average Annual Percentage Change 79-84 85-89 90-94 95-98 79-98

DK France Ger Greece Ireland -2.45 9.95 4.21 4.81 11.02 1.48 5.43 3.39 5.72 10.17 3.54 3.80 5.74 7.53 9.46 1.94 -0.76 5.35 7.40 4.35 0.91 5.14 4.62 6.24 9.08

Italy Japan Lux 5.66 3.81 1.67 2.71 6.83 1.15 5.23 9.43 2.93 4.59 8.85 7.59 4.60 6.98 3.04

28

Neth Spain Switz UK 5.92 7.23 1.70 4.04 8.19 4.98 2.25 2.42 2.74 3.79 1.12 12.74 2.20 7.13 2.00 2.07 4.95 5.79 1.75 5.41

USA 5.23 2.83 2.45 0.67 3.03


Table 6: Total Factor Productivity: 1978-98

(a) Index of TFP (1995=100) 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

DK France Ger Greece Ireland Italy 64.90 42.82 65.47 56.81 29.86 64.15 66.91 47.21 70.18 59.83 32.09 66.50 72.45 52.94 74.61 61.04 35.23 71.56 75.63 57.92 78.22 61.00 38.08 77.99 77.83 62.99 78.74 61.87 40.57 78.14 79.61 66.41 79.92 63.83 44.42 77.39 81.94 70.11 80.39 68.10 48.87 80.42 82.88 71.76 80.48 72.29 53.68 80.77 77.98 74.17 79.91 77.67 58.74 82.03 79.84 76.90 80.19 81.60 62.86 83.16 80.24 80.48 80.84 84.68 68.36 84.99 80.98 84.40 82.35 87.97 73.40 84.02 81.76 87.86 89.23 93.10 79.38 83.53 82.45 90.60 88.62 96.57 84.91 88.02 86.65 93.90 87.03 100.80 88.70 91.64 90.88 95.73 89.64 100.88 91.72 94.42 94.98 98.00 93.54 99.87 95.62 97.19 100.00 100.00 100.00 100.00 100.00 100.00 103.63 100.33 105.80 100.70 105.99 103.34 104.49 102.71 105.80 101.78 105.99 104.90 104.77 102.60 106.46 102.28 109.97 104.21

Japan 76.53 78.47 79.18 79.76 81.43 82.38 83.84 85.55 86.90 88.75 91.05 93.67 95.94 97.35 98.02 98.20 98.73 100.00 99.55 98.59 101.42

Lux 75.49 77.28 79.14 80.68 83.41 85.45 88.76 91.49 91.04 90.77 90.49 93.48 94.01 93.67 98.14 97.76 98.07 100.00 101.73 106.15 105.99

Neth 45.77 49.38 53.29 56.75 60.26 63.75 67.31 67.67 66.50 69.66 73.74 77.04 83.48 88.34 92.81 95.09 97.51 100.00 101.49 102.82 104.65

Spain 51.65 56.49 60.74 64.18 66.85 70.82 74.42 77.80 83.19 85.49 88.51 89.01 88.47 89.30 91.99 94.82 97.44 100.00 102.05 105.44 108.70

Switz 96.73 99.66 102.62 103.86 104.50 105.64 105.74 105.82 105.26 106.77 106.34 105.31 103.05 101.28 100.57 100.78 97.56 100.00 99.39 100.78 104.35

UK 63.36 64.92 67.59 68.82 69.38 70.76 73.09 74.46 75.65 76.00 76.81 78.04 82.91 90.85 95.24 97.04 98.68 100.00 100.63 101.09 114.79

USA 71.32 69.92 70.59 76.29 74.04 76.10 77.12 78.08 77.76 80.22 83.11 84.59 88.43 90.16 92.72 94.44 96.85 100.00 102.91 107.38 110.01

UK 2.42 1.32 4.84 4.00 3.07

USA 1.37 1.88 2.75 3.24 2.22

(b) Average Annual Percentage Change DK France Ger Greece Ireland 79-84 3.98 8.59 3.51 3.09 8.56 85-89 -0.19 3.78 0.49 5.26 8.48 90-94 3.26 3.04 2.65 2.60 5.45 95-98 2.50 1.16 3.33 0.60 3.58 79-98 2.46 4.51 2.50 3.01 6.77

Italy Japan Lux 3.90 1.53 2.74 0.89 2.24 1.06 2.97 1.06 0.98 1.77 0.69 1.97 2.49 1.42 1.73

29

Neth Spain Switz 6.64 6.29 1.50 2.78 3.67 -0.08 4.85 1.84 -1.51 1.78 2.77 1.71 4.26 3.82 0.40

ECONOMIC PERFORMANCE IN EUROPEAN ... - CiteSeerX

ECONOMIC PERFORMANCE IN EUROPEAN ... - CiteSeerX

Suggest Documents

ECONOMIC IMPACT IN THE EUROPEAN REGIONAL ... - CiteSeerX

Financial Liberalization and Economic Performance in ... - CiteSeerX

Cooperation and Political Economic Performance in ... - CiteSeerX

CENTRE FOR ECONOMIC PERFORMANCE ... - CiteSeerX

Economic performance of Turkey - CiteSeerX

The new European economic governance - CiteSeerX

Economic Identification - Economic Sociology_The European

Economic Performance Indicators for Fisheries - CiteSeerX

Comparative economic performance, inequality and the ... - CiteSeerX

Information Technology and Economic Performance ... - CiteSeerX

Economic Performance Indicators for Fisheries - CiteSeerX

Unionization and Economic Performance: Evidence on ... - CiteSeerX

Trade and economic performance: does Africa's ... - CiteSeerX

Unions, Dynamism, and Economic Performance - CiteSeerX

Information Technology and Economic Performance ... - CiteSeerX

Can Labour Regulation Hinder Economic Performance? - CiteSeerX

CENTRE for ECONOMIC PERFORMANCE DISCUSSION ... - CiteSeerX

Internal organization and economic performance: The ... - CiteSeerX

Fishing strategies, economic performance and ... - CiteSeerX

Political Competition and Economic Performance: Theory ... - CiteSeerX

The Financial System and Economic Performance - CiteSeerX

Governance and Economic Performance in

Economic performance and government size - European Central Bank

Economic performance and government size - European Central Bank