Telecommunications liberalisation and employment Tim Kelly and Michael Minges (December 1995) International Telecommunication Union The views expressed in this paper are the authors and do not necessarily reflect the opinion of the ITU or its Members. Outside of North America, telecommunication services have traditionally been provided by government-operated Posts, Telegraph and Telephone (PTT) enterprises. Since the early 1980s, the telecommunication service sector has witnessed dramatic restructuring. Changes include the separation of posts and telecommunications, corporatisation, privatisation, competition and foreign investment. The extent and type of change has varied world-wide but every region has been directly or indirectly affected. This paper summarises the effect these changes have had on employment patterns. Over the world as a whole, the public telecommunications service industry employs just over five million people. Employment in the public telecommunication services sector has declined by some 6 per cent since 1982 with particularly notable drops in the Pacific area (-25 per cent) and in the Americas (-23 per cent, mainly in North America). Consequently, even though telecommunications is one of the fastest growing and most profitable of all service sectors, it does not provide an answer to the structural problems of employment that are currently afflicting many of the industrialised nations. Automation of simple functions like directory enquiries or operator services, and the diminishing need for routine maintenance has reduced the sector's demand for employment. The restructuring of the sector over the last decade has tended to accelerate the decline in employment in traditional public telecommunication services because it has focused attention on cost control, productivity comparisons and the need for corporate competitiveness. There is some evidence to suggest that the employment decline in public services has been partially offset by an increase in employment in the non-traditional telecommunication sector (e.g., companies providing Value-Added Network Services, capacity resale, closed-user group networks, consultancy and training services, etc.) and in private networks (e.g. intra-corporate networks). For instance, it is estimated that the mobile communications industry worldwide (equipment and services) employs more than a quarter of a million people. These employees are in a sector which did not exist ten years ago. However, they hardly replace the jobs lost in the traditional sector. The restructuring of the telecommunication service sector has tended to follow a specific sequence. First, post and telecommunication operations are separated. This is typically followed by corporatisation of the telecommunication operator. Then, the operator is partly or fully privatised. Finally, competition may be introduced. Each of these stages appears to have a different effect on employment patterns often with the impact on the telecommunication organisation's employment level preceding the actual restructuring stage. The remainder of the paper focuses on specific region and country experiences. Table 1. Telecommunications employment and productivity Regional totals and percentage change, public telecommunication services sector, 1982-92 Region

Africa Americas Asia Europe Oceania Former-USSR

Telecom employment, (000s) 1992 233 1'090 1'643 1'277 87 689

Change in employment, 1982-92 (%) -1% -23% 9% -5% -25% n.a.

Mainlines per employee, 1992 40 177 72 152 114 59

Change in mainlines per employee, 1982-92 (%) 9% 36% -1% 10% 37% n.a.

World 5'019 -6% 112 12% Note: Percentage change calculations based on totals for only those countries for which data is available in both the start year (1981/82) and the end year (1991/92). Source: ITU/BDT Telecommunication Indicator Database.

1

The British experience The UK was the first European country to introduce competition for telecommunication services when Mercury Communications Limited was granted a license in 1982 to compete with British Telecommunications plc (BT), the existing service provider. BT, which was established as a public corporation in 1981 to operate telecommunication services previously provided by the General Post Office, was gradually sold and by 1993 it was 99 per cent privately owned. In 1991, the duopoly was abolished and market entry is effectively completely open for most domestic telecommunication services, though a licensing system has been retained. Cable television companies, many of which are licensed to provide telephony services, have been among the most successful of the new market entrants in terms of attracting new subscribers with just over 300,000 subscribers in total. There are also four major suppliers of mobile communication (Vodafone, Cellnet (60% owned by BT), Mercury One-to-One and Orange) which together have more than two million subscribers. BT inherited a large government bureaucracy from the General Post Office and initially began shedding staff. Over 10,000 employees were released between 1982 and 1987. However this was reversed in 1987 when it became clear that job shedding was having an adverse effect on service quality and routine maintenance operations which had been exacerbated by an industrial dispute. Another reason was the need to convert BT's exchange lines to digital technology. In 1986, less than 2 per cent of BT's local exchange lines were connected to digital exchanges putting it at a serious disadvantage with competitor Mercury which was building an all digital network. The rapid replacement of BT's exchanges from analogue to digital required additional manpower.1 By 1989, staff levels were back to where they had been in 1982. But beginning in 1989, BT once again began reducing staff. Almost 90,000 jobs were lost from 19891993, a 37 per cent reduction. Further reductions have been announced, particularly among middle management and administrative staff, which would cut a further 50'000 jobs and should bring BT into line with the levels of labour productivity (as measured by main lines per employee) achieved by the Regional Bell Operating Companies in the United States. In addition, BT is increasingly using outsourcing and subcontracting of work rather than awarding direct employment contracts.

Figure 1. Modernisation, competition and employment in the UK BT employees and digital exchange lines, 1986-93 and total telecom service employment, 1989-93 BT

80%

Employees (000's)

200

60% 50%

150

40% 100

30%

50

20%

Employees

10%

Digital lines

BT Mercury

250

Vodafone 200 150 100

0

1993

1992

1991

1990

1989

1988

1987

Telecom service employment in the UK

50

0% 1986

0

Main lines connected to digital exchanges

70%

300 Employees (000's)

250

1989

1990

1991

1992

1993

Note: Years beginning April l. Source: ITU/BDT PTO database.

Increased competition from Mercury and Vodafone and the licensing of cable television operators to provide telephone services have influenced BT's need to increase productivity. However, there is evidence, that BT's productivity problems preceded competition. For example, in 1982, BT had the lowest productivity of any European Union operator except Portugal's CTT. Though productivity has risen, it is still below the European Union average and well below Mercury's. BT would have to shed 80,000 workers to reach Mercury's productivity level (in terms of revenue per staff) and over 20,000 to reach the European Union average of 178 lines per employee. Though employment in Mercury and Vodafone is growing, it is not close to making up for the amount being dropped by BT. For example, although Mercury and Vodafone added over 1,000 employees from 1991 to 1992, BT shed almost 40,000. Employment figures for the nascent cable television / telephony industry were unavailable at the time this report was drafted but it is likely that these are more than 5'000 employees.

1 Ironically this change would result in a future need for less staff since digital exchanges require less maintenance and service intervention. 2

Table 2. Telecommunication employment and productivity in the UK Employees and revenue in UK, 1991-1992 Employees Revenue (£m) Revenue per employee 1991-92 change 1991 1992 Total % 1991 1992 1991 1992 BT 210’500 170'700 -39'800 -18.9% 13'337 13'242 63'359 77'575 Mercury 6’369 7'259 890 14.0% 802 1'062 125'860 146'287 Vodafone 2’142 2'333 191 8.9% 585 664 273'267 284'655 UK 216’869 177'959 -38'910 -17.9% 14'724 14'968 67'893 84'109 Note: Year beginning April 1. Not including Kingston Communications or cable television / telephony companies. Source: ITU/BDT PTO database.

Other European Union Though BT has shed the most staff -- in both volume and percentage terms -- of European Union telecommunication operators, employment throughout the Union has been declining. The status of the operator (e.g., private, partly private or government-owned) appears to make little difference. State-owned operators such as Greece's OTE or France Télécom have reduced staff while partly private operators such as Italy's SIP or Spain's Telefónica have increased staff. This suggests that a more critical factor is competition, or the threat of it, which is not at the same level in the rest of the European Union as in the UK. Table 3. Employment and productivity in the European Union Main lines, employees and main lines per employee, European Union, 1988-92 Main lines Employees Main lines per employee 1988 1992 CAGR Change CAGR 1988 1992 CAGR CAGR 000s 000s 1988-92 1988 1992 1988-92 1988-92 1988-92 1988-92* Belgacom 3'525 4'264 4.9% 26'331 26'843 512 0.5% 134 159 4.4% 4.4% BT (UK) 23'740 26'084 2.4% 244'400 170'700 -73'700 -8.6% 97 153 12.0% 3.4% CTT (Portugal) 821 1'581 17.8% 10'291 11'600 1'309 3.0% 80 136 14.3% 14.3% DBP Telekom 27'823 35'420 6.2% 216'000 231'000 15'000 1.7% 129 153 4.5% 4.5% Dutch PTT 6'466 7'395 3.4% 29'142 34'949 5'807 4.6% 222 212 -1.2% -1.2% France Telecom 25'827 30'100 3.9% 159'520 155'300 -4'220 -0.7% 162 194 4.6% 3.9% Luxembourg P&T 167 206 5.4% 745 771 26 0.9% 224 267 4.5% 4.5% OTE (Greece) 3'618 4'496 5.6% 30'327 26'716 -3'611 -3.1% 119 168 9.0% 5.9% SIP (Italy) a) 20'092 23'709 4.2% 80'862 87'475 6'613 2.0% 248 271 2.2% 2.2% Telecom Eireann 842 1'113 7.2% 14'270 13'033 -1'237 -2.2% 59 85 9.7% 7.4% TeleDanmark 2'791 3'002 1.8% 18'720 17'890 -830 -1.1% 149 168 3.0% 1.9% Telefónica de España 10'971 13'792 5.9% 66'062 74'437 8'375 3.0% 166 185 2.8% 2.8% EU 126'686 151'163 4.5% 896'670 850'714 -45'956 -1.3% 141 178 5.9% 4.6% Note: For Portugal and UK only largest wire-line telephone companies shown. Data for BT and Telecom Eireann refer to year beginning April 1. a) SIP only provides local exchange and mobile services so its productivity figures will be higher than "full-service" operators that provide long distance services. * After deducting employment losses. Source: ITU/BDT PTO database.

3

The Latin American experience Latin America has been the scene of a number of recent telecommunication privatisations. The privatisations have been characterised by sales to large foreign telecommunication operators rather than by sales to the public which occurred in the UK. The impact of this form of privatisation on employment is mixed. For example, in five countries where privatisation occurred, employment dropped in three and rose in the other two. In countries which have not been privatised, employment has generally been declining. However, because the newly-private operators have been increasing the size of their network at a much faster rate than the state-owned companies, their rate of labour productivity growth has also been correspondingly higher. Table 4. Employment and productivity in Latin America Main lines and employees in selected Latin American countries, 1990, 1992 Main telephone lines 1990 1992 000s 000s CTC (Chile) 811 1'213 CANTV (Ven.) 1'494 1'804 Telecom Argentina 1'391 1'673 Telefónica de Argen. 1'695 2'008 Telmex 5'189 6'545 Private 9'771 12'032 ENTEL-Peru 221 231 TELEBRAS 8'500 9'900 CPT Peru 352 381 Antel Uruguay 415 492 ICE (Costa Rica) 267 309 State 9'756 11'314

Employees CAGR 1990-92 49.4% 20.7% 20.3% 18.5% 26.1% 23.1% 4.8% 16.5% 8.3% 18.5% 15.8% 16.0%

1990 7'530 19'861 19'002 21'770 49'912 110'545 9'904 93'100 5'952 6'730 3'267 118'953

Main lines per employee 1992 7'991 22'594 17'041 19'252 48'937 107'824 7'506 89'600 5'156 6'201 3'325 111'788

Change 1990-92 461 2'733 -1'961 -2'518 -975 -2'721 -2'398 -3'500 -796 -529 58 -7'165

CAGR 1990-92 6.1% 13.8% -10.3% -11.6% -2.0% -2.5% -24.2% -3.8% -13.4% -7.9% 1.8% -6.0%

1990 108 75 73 78 104 88 22 91 59 62 82 82

1992 152 80 98 104 134 112 31 110 74 79 93 101

CAGR 1990-92 40.8% 6.1% 34.1% 34.0% 28.6% 26.2% 38.3% 21.0% 25.0% 28.6% 13.8% 23.4%

Note:

Operators were privatised in Chile in 1988, in Argentina and Mexico in 1990 and in Venezuela in 1991. The Peruvian operators were privatised in 1994. Source: ITU/BDT PTO database.

The Asia-Pacific experience Telecommunication employment patterns in the Asia and Pacific region have differed depending on the level of economic development. The three member countries of the Organisation for Economic Co-operation and Development (OECD) in the region have seen overall telecommunication employment fall. Developing Asia presents an interesting contrast for it is the one region in the world where telecommunication employment is not dropping. A primary reason is the buoyant state of the regional economy which is driving demand for telecommunication services. Trends for operators in Korea, Malaysia, Singapore and Thailand, shown in Table 5, reflect this where employment increased 2.3 per cent a year between 1988-92. At the end of 1992, all were state-owned operators except Telekom Malaysia which became partly private in 1990 (some shares in Singapore Telecom and Korea Telecom were sold in 1993). By the end of 1992, productivity in these countries was well above the world average and close to the European average showing that state-owned operators can be as productive as private ones. In fact Korea Telecom has the third highest productivity in the world (after Italy's SIP and Luxembourg Post and Telephone enterprise). The case of the Telephone Organisation of Thailand (TOT) shows that growing Build-Operate-Transfer schemes in that country have had not decreased employment for incumbent operators. Table 5. Employment and productivity in Asia Main lines and employees in East Asia, 1988, 1992 Main lines 1988

Employees CAGR 1988 1992 1988-92 Korea Telecom 10'306'028 15'593'454 10.9% 52'073 59'137 Singapore Telecom 938'719 1'169'089 5.6% 11'620 10'856 Telekom Malaysia 1'247'687 2'091'578 13.8% 28'168 29'577 TOT (Thailand) 1'005'872 1'790'029 15.5% 17'956 20'855 13'498'306 20'644'150 11.2% 109'817 120'425 Note: Korea Telecom and Singapore Telecom partly privatised in 1993, Telekom corporation that franchises certain telecommunication operations. Source: ITU/BDT PTO database. 1992

4

Main lines per employee Change CAGR 1988 1992 CAGR 1988-92 1988-92 1988-92 7'064 3.2% 198 264 7.4% -764 -1.7% 81 108 7.5% 1'409 1.2% 44 71 12.4% 2'899 3.8% 56 86 11.3% 10'608 2.3% 123 171 8.7% Malaysia partly privatised in 1990. TOT is a state-owned

The situation has been different for the OECD countries where overall employment has been falling. In Japan, competition was introduced in 1985 and the incumbent domestic operator, Nippon Telephone and Telegraph Corporation (NTT) was partly privatised. Since then, NTT has shed 55,000 staff, almost 20 per cent of its workforce. As in the UK, the new competitors have not come close to replacing the jobs lost by NTT, though market liberalisation has allowed some gains among the thousand or so service providers (Type II carriers) that offer value added services and other enhanced services over facilities provided by NTT and the new common carriers. In Australia, the telecommunication sector also underwent significant restructuring. The incumbent operator, now called Telstra, reduced its staff level by almost 15,000 or 15 per cent in 1992. The new competitor, Optus Communications, has around 2,000 staff. The Telecom Corporation of New Zealand presents an interesting case of the effect on employment of sector changes over a condensed period of time. Telecom was created from the Post Office in 1987, privatised in September 1990 and began facing competition in 1991 (from CLEAR Communications Limited). It has shed almost half the work force it inherited from the Post Office: staff levels went from 23,900 in March 1988 to 12,338 in March 1993. CLEAR had 475 staff in July 1993; again it seems that the new market entrants are not making up for the staff being lost by Telecom.

Figure 2. Corporatisation, privatisation and competition in New Zealand Staff of Telecom Corporation of New Zealand, 1987-93 Main lines per employee

25'000

Employees Main lines per employee

160

20'000

0

Competition

5'000

Privatised

10'000

Corporatised

15'000

140 120

Actual staff level 1987 staff level

100 80 60 40 20 0

1987 1988 1989 1990 1991 1992 1993

1987 1988 1989 1990 1991 1992 1993 Note: Years beginning April 1. Source: ITU/BDT PTO database.

The US experience In the United States, competition for the provision of long distance was liberalised in 1984 and AT&T's local exchange operations were divested. Since then, employment by the local exchange carriers has dropped by over 150,000 reversing a thirty year increase in employment in the industry. By 1992, employment levels were back to where they had been in 1950. Ironically, except for cellular services, the local exchange carriers face little direct competition in most States. The two main long distance competitors to AT&T, Sprint and MCI have added almost 20,000 employees since 1988. However this does not make up for the overall decline during that period of about 70,000. Statistics are not available for other long-distance and Value-Added Network providers who may have increased employment, though, as in the UK, the total amount is believed to be quite low relative to the loss of jobs in the main local exchange carriers. Though statistics for the cellular industry show a net gain of employment of about 23,000 since 1988, this figure is subject to double counting since the local exchange carriers include cellular staff in their overall employee statistics. AT&T, which has faced the most competition, shed over 7,000 jobs.

5

Figure 3. Employment trends in the US Telecommunication service employment, US, 1950-1990, thousands

1200

1200

1000

1000

800

800

600

600

400

400

200

200

0 1950

0 1955

1960

1965

1970

1975

1980

1981

1983

1985

1987

1989

Source: Left chart: FCC. Right chart: USTA.

Table 6. Employment and productivity in the US Main lines

RBOC GTE RBOC & GTE Sprint MCI AT&T Long distance Total

1988 (000s) 99'535 14'504 114'039

1992 (000s) 110,257 16,819 127,076

Employees CAGR 1988-92 2.6% 3.8% 2.7%

1988 560,775 98,000 658,775 37,700 17,596 85,531 140,827 799,602

1992 495,247 81,053 576,300 43,400 30,964 78,387 152,751 729,051

Change 1988-92 -65,528 -16,947 -82,475 5,700 13,368 -7,144 11,924 -70,551

Employees per line CAGR 1988-92 -3.1% -4.6% -3.3% 3.6% 15.2% -2.2% 2.1% -2.3%

1988 177 148 173

1992 223 208 221

CAGR 1988-92 5.8% 8.8% 6.2%

Note: RBOC = Regional Bell Operating Companies. Along with GTE, they provide majority of local telephone service. Source: ITU/BDT PTO database.

Conclusions Employment in the world-wide telecommunication sector is dropping regardless of region or operator status (e.g., government run, state-owned, partly private, or fully private). While the introduction of competition often results in new operators providing employment opportunities, the amount of new staff hired is substantially below the levels released by incumbent operators. There is no example of an overall net increase in employment as a direct result of competition. The impact of privatisation on telecommunication employment is mixed. In some countries it has resulted in a drop in staff while in others it has resulted in an increase in staff. The statistics do not show that privatisation by itself results in relatively higher productivity growth in the telecommunications sector. There is some evidence to suggest that employment losses precede the introduction of competition or privatisation in the sense that the operator is being "prepared" for a change in its status.. Future directions In order to carry the debate on employment trends further, it will be necessary to conduct more research on specific issues. One area for improvement is in the comparability of staff indicators. For example, comparisons are affected by the inclusion of postal or overseas staff. Also some operators contract out installation work which will tend to overstate staff per telephone line. Additional productivity measures need to be developed. Staff per line is distorted by telephoneonly companies as opposed to companies that also provide telex and other services; staff per line will be overstated for the former. Also, staff per line is meaningless for long distance only operators. Financial productivity measures (e.g., revenues per staff) are often a reflection of differing income levels rather than output. A further area for research is to investigate whether market liberalisation has created new employment opportunities outside the traditional carriers. These new positions are as likely to be found in the employment of telecommunication users, which now plan and manage their own networks, as among the traditional PTOs and third party service providers. For this reason, research may need to be based on occupational data rather than on enterprise-level data. Finally, it will be necessary to take a more sophisticated approach to measuring productivity. The best measure of the "output" of a PTO is not necessarily

6

the number of lines or even the value of sales, but rather the volume of traffic. At present, few operators report sufficiently detailed traffic statistics to allow for a calculation of productivity growth in terms of traffic per employee.

7