Constraints to Growth and Employment in South Africa20112351832

Document Sample
Constraints to Growth and Employment in South Africa20112351832 Powered By Docstoc
					                                       DISCUSSION PAPER




              Constraints to
         Growth and Employment
             in South Africa
  REPORT NO.1: STATISTICS FROM THE LARGE MANUFACTURING FIRM SURVEY


                                                 Vandana Chandra
                                                      Lalita Moorty
                                                  Bala Rajaratnam
                                                  Kendall Schaefer




                                                        June 2001



THE WORLD BANK                          INFORMAL DISCUSSION PAPERS
SOUTHERN AFRICA                                  ON ASPECTS OF THE
DEPARTMENT                                ECONOMY OF SOUTH AFRICA
                   PREVIOUS WORLD BANK PAPERS ON SOUTH AFRICA


Previously published in the World Bank series of informal discussion papers on South Africa:

Levy, B. January 1992. "How Can South African Manufacturing Efficiently Create Employment? An analysis of
the Impact of Trade and Industrial Policy".

Kahn, B., Abdel, S. and Walton, M. May 1992. "South Africa: Macroeconomic Issues for the Transition".

Fallon, P. October 1992. "An Analysis of Employment and Wage Behavior in South Africa".

Southern Africa Department, May 1993. "An Economic Perspective of South Africa".

Belli, P., Finger, F., Ballivian A., August 1993. "South Africa" A Review of Trade Policies".

Riley, T., November 1993. "Characteristics of and Constraints Facing Black Businesses in South Africa: Survey
Results".

Southern Africa Department, February 1994. "South African Agriculture: Structure, Performance and Options for
the Future".

Fallon, P., Pereira da Silva, L., April 1994. "South Africa: Economic Performance and Policies".

Southern Africa Department, April 1994. "South Africa: Financing the Metropolitan Areas of South Africa".

Southern Africa Department, June 1994. "Reducing Poverty in South Africa".

Southern Africa Department, December 1995. "South Africa: Natural Resource Issues in Environmental Policy".

Southern Africa Department, December 1995. "South Africa: Education Sector Strategic Issues and Policy
Options".

Levy, B., February 1996. "Southern Africa: The Business environment for Industrial Small and Medium
Enterprises".

Fallon, P., Lucas, R., November 1998 (revised). “South Africa Labor Markets Adjustment and Inequalities”.

Tsikata, Yvonne M., July 1999. ”Liberalization and Trade Performance in South Africa”.

Chandra, V., Moorty, L.,Rajaratnam, B., Schaefer, K., June 2001. “ Constraints to Growth and Employment in
South Africa, Report No. 1: Statistics from the Large Manufacturing Firm Survey”.


In addition, a number of technical and seminar papers prepared by World Bank staff and South African
counterparts in key sectors have been discussed in the country.
   CONSTRAINT S TO GROWTH AND EMPLOYMENT IN
                  SOUTH AFRICA
     REPORT NO.1: STATISTICS FROM THE LARGE M ANUFACTURING FIRM S URVEY




                                    Vandana Chandra
                                     Lalita Moorty
                                    Bala Rajaratnam
                                    Kendall Schaefer




The findings, interpretations and conclusions expressed in this report are entirely those of the
authors. They do not necessarily represent the views of the World Bank Group, its Executive
Directors, or the countries that they represent and should not be attributed to them.
                                                             CONTENTS

Foreword.......................................................................................................i
Executive summary.....................................................................................iii
1. Introduction...........................................................................................1
  Motivation and objectives ..................................................................................................... 1
  Sample................................................................................................................................... 2
  Sample characteristics........................................................................................................... 3
  Organization of the report..................................................................................................... 5
2. Constraints to Business Growth ............................................................6
  International comparisons ................................................................................................... 10
3. Location Constraints ...........................................................................12
  Crime and violence in GJMA ............................................................................................. 12
  Infrastructure and service delivery in GJMA...................................................................... 13
  Choice of Location for Expansion...................................................................................... 16
4. Corporate Finance...............................................................................17
  Sources of financing............................................................................................................ 17
  Firms’ responses to the interest rate hike............................................................................ 19
  Other firm characteristics.................................................................................................... 20
5. Exchange Rate and Trade Policy.........................................................22
  Trading profile of firms ....................................................................................................... 22
  Exchange rate shocks and export behavior......................................................................... 23
  Competitors in global markets ............................................................................................ 24
  Excha nge rates and forward contracts ................................................................................ 25
  Trade policy ........................................................................................................................ 27
6. Labor Regulations and Implicit Costs.................................................31
  Role of Large Manufacturing firms in employment generation......................................... 31
  Firm response to recent labor regulations ........................................................................... 32
  Labor market environment.................................................................................................. 33
  Temporary work and subcontracting .................................................................................. 37
7. Availability of Skills.............................................................................40
  Racial distribution of skills employed in GJMA ................................................................ 40
  Difficulty in finding skills ................................................................................................... 40
  Vacancy rates and levels ..................................................................................................... 41
  Recruitment and hiring preferences .................................................................................... 42
  Worker training ................................................................................................................... 43
  Type and magnitude of training .......................................................................................... 44
  Do trained workers leave after training?............................................................................. 45
8. Licensing Regulation and Corruption .................................................46
  Start- up licenses and permits .............................................................................................. 46
  Licensing and permits to continue operations .................................................................... 46
  Cumbersome procedures and corruption ............................................................................ 47
  Government contracts ......................................................................................................... 48
Annex 1. Local Economic Development Methodology...............................49
Annex 2. Sample Design and Weights........................................................51
Annex 2. Sample Design and Weights........................................................52
Annex 3. Demographics .............................................................................55
Annex 4. Constraints to Business Growth..................................................59
Annex 5. Crime and Infrastructure in GJMA...........................................60
Annex 6. Exchange Rate and Trade Policy................................................62
Annex 7. Employment Profile in GJMA....................................................64
Annex 8. Labor Regulations.......................................................................67
Annex 9. Availability of Skills....................................................................69



This report was prepared by Vandana Chandra (Task Manager), Lalita Moorty, Bala Rajaratnam, and Kendall
Schaefer. We are indebted to Junaid Ahmad (AFCO1) and Ketso Gordhan (City Manager, GJMC) for
initiating the LED Project. We gratefully acknowledge help from our peer reviewers: T. Biggs (AFTPS), M.
C. Hallward-Driemeier (DECRG), B. Levy (AFTI2), and A. Stone (PSAFB). We also received helpful
comments from P. Cox (AFRVP), A. Gelb (AFTMI), P. Le Houerou, J. D. Lewis, J. Sackey, L. Pereira de
Silva, F. Yagci (AFTM1), and G. Batra (PSAFB). In South Africa, we received financial and logistical
support from M. Maganlal, R. Seedat, and B. Shibambu (Strategic Planning Unit, GJMC). A special note of
thanks to Professor J. Martins, and D. Tustin (BMR) for expert survey management and to the survey teams in
South Africa who were drawn from RAU, University of Pretoria, UNISA, Wits, Vista and the BMR staff. We
                                                                               nd
also thank B. Smit (BER), I. Macun (DOL), C. Rogerson, R. Tomlinson, a hosts and participants of
seminars held at GJMC, DTI, Durban Metro Council, University of Cape Town, DPRU, University of
Stellenbosch and Urban Forum (Washington). The views expressed here do not reflect those of the World
Bank or its members and we accept responsibility for any remaining errors and omissions.
                                          LIST OF BOXES

Box 1.1:     Macroeconomic background                                                       2
Box 5.1:     Average proportion of manufacturing output exported in East Asia in 1997-98   22
Box 5.2:     Export dynamics and productivity in Mexico                                    24
Box 5.3:     Facilitating industrial clusters in Morocco                                   30

                                         LIST OF CHARTS

Chart 1.1:   Distribution of full-time manufacturing employment in GJMA, 1998               4
Chart 1.2:   Age of large manufacturing firms in South Africa, 1999                         4
Chart 1.3:   PDI participation in Large Manufacturing Firms, 1999                           5
Chart 2.1:   Index of CEO rankings on constraints to growth                                 6
Chart 3.1:   Types of crime in GJMA, 1998                                                  12
Chart 3.2:   Firm ratings of location on proximity to infrastructure and services, 1999    14
Chart 3.3:   Firm ratings of availability of services, 1999                                15
Chart 3.4:   Firm rating of location on cost and availability of services, 1999            15
Chart 3.5:   Choice of location for expansion                                              16
Chart 4.1:   Percent of investment capital by size-class,1998-99                           18
Chart 4.2:   Firm responses to interest rate hike 1998                                     19
Chart 5.1:   Firm response to depreciation, 1998                                           23
Chart 5.2:   Forward contracts on foreign currency exposure, 1998                          26
Chart 5.3:   Comparing response to currency depreciation, 1998                             26
Chart 5.4:   Firm ratings of trade barriers, 1999                                          28
Chart 5.5:   Awareness and use of special facilities for exports, 1999                     29
Chart 5.6:   Importance of special incentives for exports                                  29
Chart 5.7:   Firm ratings of costs of special incentives for exports                       30
Chart 6.1:   Ratio of non-managerial workers to managers per firm                          31
Chart 6.2:   Response of firms in GJMA to individual recent labor regulations, 1998        32
Chart 6.3:   Cumulative response of firms in GJMA to all recent labor regulations, 1998    33
Chart 6.4:   Number of unions GJMA’s large manufacturing firms work with                   34
Chart 6.5:   Level at which collective agreements are made                                 35
Chart 6.6:   Hiring and firing costs of an entry level worker, 1998                        37
Chart 6.7:   Use of temporary workers and subcontractors, 1998                             38
Chart 6.8:   Reasons for using temporary workers, 1998                                     38
Chart 6.9:   Tasks outsourced by firms, 1998                                               39
Chart 6.10   Reasons for subcontracting production activity, 1998                          39
Chart 7.1:   Distribution of skills by race in GJMA, 1998                                  40
Chart 7.2:   Difficulty in finding skills, 1999                                            41
Chart 7.3:   Recruitment preference for production workers in GJMA                         43
Chart 8.1:   Firms that found government procedures cumbersome                             47

                                          LIST OF MAPS

Map 1.1:     Industrial zones and large firm sample                                         3
Map 4.1:     Firms that increased investment between 1997 and 1998                         21
Map 4.2:     Firms that increased both investment and employment between 1997 and 1998     21
                                        LIST OF TABLES

Table 2.1:   Rankings of constraints to business growth in 5 East Asian countries         10
Table 2.2:   Constraints to business growth in Latin America, OECD, and South Africa      11
Table 3.1:   Firm spending on crime prevention, 1998                                     13
Table 3.2:   Correlation of expenditure on crime prevention with economic variables      13
Table 5.1:   Exports by destination                                                      23
Table 5.2:   Top five competitors in key regional export markets                         25
Table 6.1:   Number of strikes, 1998                                                     36
Table 6.2:   Number of workdays lost in GJMA due to strikes, 1998                        36
Table 6.3:   Time taken to retrench an entry level worker                                37
Table 7.1:   Worker training by GJMA firms, 1998                                         45
Table 8.1:   Firms that found regulations costly for starting a business                 46
Table 8.2:   Firms that found regulations very costly for continuing operations          47
Table 8.3:   Median number of government tenders submitted and approved, 1997 and 1998   48
                                                   -i-




                                         FOREWORD
In an urbanized context like South Africa, the role of cities in economic growth and poverty reduction
is of national interest. In South Africa, over 80 percent of national GDP is contributed by the urban
sector and about 60 percent of the population resides in urban conurbation. Nowhere is the urban
economic and demographic concentration more visible than in South Africa’s large metropolitan
areas – Cape Town, Durban, Pretoria, East Rand and Johannesburg – which together account for 35
percent of the national economic pie and an equally large portion of the national population. In South
Africa, therefore, the cities and the nation share a destiny that is intertwined. The policy question that
naturally arises in this context is what are the mechanisms through which a metropolitan center can
contribute to national targets of growth and poverty reduction?

       Traditionally, the answer has focused on mechanisms of fiscal stability, provision of local
infrastructure and market efficiency. To begin with, a city capable of raising its own revenues and
financing its own needs, contributes to the national objective of macro fiscal stability, a prerequisite
for economic growth and poverty reduction. Similarly, a city capable of providing municipal
infrastructure in an efficient and equitable manner contributes directly to the national economy.
Households and businesses receiving reliable and affordable services can in turn meet the
consumption and production needs of the economy. In this context, cities that select innovative
partnerships in the financing and delivery of infrastructure by involving public-private-and
community organizations act as national pilots of innovation. Such partnerships are often an
important source of employment generation and direct foreign investment. Finally, by making sure
that urban markets for land and transport are not impeded by local policies, cities establish the micro
foundations for economic growth by permitting, for example, housing to be delivered more
effectively or by facilitating the mobility of households.

       In taking forward the iGoli 2010 framework, Johannesburg has adapted and innovated on the
global best practices of fiscal management, infrastructure delivery and making basic municipal
markets work. The iGoli framework for better economic management of cities is thus Johannesburg’s
response to the divisions and dysfunction of the apartheid city. IGoli is also Johannesburg’s
contribution to national goals of economic growth and poverty.

       But, in formulating the IGoli plan, the Johannesburg leadership also asked whether in addition
to the routes of fiscal sustainability, infrastructure provision, and efficient markets, the city could
directly influence employment generation. Cognizant of the trap of trying to pick “winners and
losers” in the economic sphere, Council is seeking to assess its potential role in employment
generation by understanding the urban economy of Johannesburg and its dynamics. The underlying
philosophy of the approach was simple. Systematic and complete information was needed to gain an
understanding of the pulse of Johannesburg’s economy and, in particular, to assess the critical barriers
to the growth of formal and informal economic nodes. Only this form of an information base –
comprehensive and in depth in nature -- would allow the decision-makers to identify the areas of
policy intervention.

       Equally important, the information system would also allow the policy makers to understand
whether the impediments to growth and employment generation and the possible policy solutions
                                                    t
were national or local in nature. In other words i would delineate the roles of national and local
governments in meeting the challenges of economic growth and employment generation. This is an
important point. Too often it has been assumed that macro instruments of fiscal and monetary control
are the main levers of economic growth. Yet, in reality, these are necessary but not sufficient policy
instruments. Intervention at the micro level in the operation of labor, land, and capital markets may
                                                  -ii-

also be needed in conjunction with macro level instruments. With cities playing such a national role
in South Africa’s economic structure, economic analysis at the city level is therefore a pre-requisite
to understanding the microeconomics of macro growth strategies. In addition, with Johannesburg
representing 42 percent of national manufacturing firms, the analysis even at the Johannesburg level
by itself will provide important insights for national level policy makers.

       The comprehensive economic information base was established under the joint partnership of a
team comprising of staff members from the City of Johannesburg and the World Bank. The task team
leader from the City’s side was Mr. Rasheed Seedat and from the World Bank side, Dr. Vandana
Chandra. Primarily drawing on local expertise and in collaboration with the Bureau of Market
Research, UNISA, the study undertook six primary surveys. To its credit, the team trained local
students from all corners of Johannesburg including Soweto, Alex, and the Northern Suburbs. These
students undertook the survey under supervision of senior staff from the World Bank and BMR. The
growth of local capacity in the area of urban economic surveying was itself a very valuable
contribution of the exercise. As a result, the work was completed in record time – design,
implementation of the surveys and data cleaning all happened in less than 12 months – and at record
cost savings.

       The information system will also serve as the base of the city’s “Economic Intelligence Unit.”
Council was cognizant that information of the form being collected in the surveys will need to be
updated periodically, analyzed continuously, and the assessments will need to be fed into the policy
process in real time. The development of such institutional capacity was also an important
achievement of the survey process. The data is available to the public through the City’s website and
will be updated automatically as new data is entered over time. Council’s intention is to honor its
commitment to transparency and to invite debate and analysis of its policies based on actual data
collected in real time.

        Finally, through the surveys a methodology was devised to enable Council to track the impact
of its expenditure patterns on access to services by the poor. Poverty monitoring is analytically and
logistically a very demanding task. Yet, for policy makers dedicated to the improvement of the
welfare of the poor – a commitment that drives the Council’s policy process – it was important that
ultimately the City is capable of assessing the impact of its polices on the access by the poor of
municipal services. But, such tracking would have to be undertaken in real time to inform the policy
process regularly. Reliance on annual household surveys did not seem to respond to this need. The
team rose to this challenge and has contributed a process that is analytically stringent and logistically
manageable and will support the Council’s objective of assessing access to services.

       Ultimately such a comprehensive and ambitious task could not have been undertaken without
the collaboration and support of many organizations. We would like to thank Agence Fraincaise
Development Propario (AFD) and Inca-French Fund for financial support, NBI and JCCI for
logistical support with surveys and most importantly, CEOs of firms, the real participants of the
survey, for their time and dedication to their City.


Ketso Gordhan, City Manager,
Greater Johannesburg Metropolitan Council, South Africa

Junaid Ahmad, Deputy Resident Representative
World Bank Resident Mission Office, South Africa

August 2000
                                                    -iii-


                                  EXECUTIVE SUMMARY
South Africa’s macroeconomic performance was characterized by low growth and persistent
job shedding in the 1990s. To understand the micro- foundations of this weak macroeconomic
performance, the World Bank, in partnership with the Greater Johannesburg Metropolitan
Council (GJMC), conducted a series of firm surveys based on the theme of Local Economic
Development (LED). This report, first in a series, 1 describes constraints to growth and
investment as reported by large manufacturing firms 2 located in the Greater Johannesburg
Metropolitan Area (GJMA) and has implications for the larger South African economy.

      Our primary objective is to examine factors that constrain firm growth and
employment in GJMA and in the national urban economy. In addition, we seek to identify
key areas where efforts can be made to improve the business environment in which firms
operate. Towards this end, using 7 exhaustive questionnaires, we interviewed the CEOs and
senior mana gers of 325 large manufacturing firms spanning eight sectors. The first
questionnaire was used to elicit CEO perceptions on the prevailing investment climate and
future prospects. This was augmented by detailed thematic questionnaires administered to
senior managers to see how the performance and operations of firms were affected by the
constraints identified by CEOs.

       Interviews with firm CEOs and senior managers revealed that while marginal gains
may result from fine-tuning macroeconomic policies, these efforts need to be augmented by
easing the structural constraints to growth. Our analysis of firm responses reveals that crime
and violence along with a shortage of skilled labor are the two main impediments that large
manufacturing firms have to contend with. Survey results relating to impact of labor
regulations on firms were inconclusive. We also found that the channels through which
movements in interest rates and exchange rates affect firms’ performance and operations are
not straightforward.

• Crime and violence were identified as the main constraints to investment and growth by
95 percent of CEOs of large manufacturing firms. Reflecting the severity of the problem,
follow-up interviews with senior managers revealed that as many as 83 percent of the firms
were victims of several types of crime in 1998. In addition, 61 percent of firms surveyed
mentioned that their employees were victimized on their way to or from work. On average,
we found that firms spent 1.6 percent of their sales revenue on security guards and devices in
1998. Approximately 50 percent of the firms reported increasing their expenditure on crime
prevention between 1997 and 1998.

• Interviews with senior managers also revealed that shortage of skilled labor is a
bottleneck to firm growth and investment. Approximately 80 percent of firms reported
extreme to moderate difficulty in finding people with managerial and professional skills.
Around 70 percent of firms reported extreme to moderate difficulty in finding people with

1
  We also surveyed firms in the large services sector, small, medium and micro enterprises (SMMEs), informal
sectors firms, informal households, training and credit providers and production-level workers in GJMA. The
issue of non-income poverty monitoring is addressed separately.
2
    Firms with more than 50 full-time employees.
                                              -iv-


service and craft skills. If finding people with appropriate skills was difficult in a year when
the economy was sluggish, it is likely that skill shortages could become a binding constraint
when the economy begins to grow at a rapid pace. Despite the fact that firms f und it     o
difficult to find skilled workers, we found that only 35 to 45 percent of firms surveyed
provide training to approximately a third of their workers.

• The impact of labor regulations on firm operations is unclear. Although CEOs perceived
labor regulations as an important constraint to firm growth, approximately 60 percent of
senior managers reported that all recent labor regulations taken together had no cumulative
impact on their employment decisions. The remaining 40 percent of the firms reported hiring
fewer workers, substituting labor with capital, and hiring temporary workers in response to
the recent labor regulations. We found some evidence to show that these regulations, aimed
at redressing discriminatory policies of the apartheid past, may have unintentionally resulted
in lowering the flexibility of firms to respond to changing market conditions.

• Although 61 percent reported being adversely affected by interest rates, the responses of
                                                            n
senior managers suggest that a significant increase in i vestment may not take place in
response to a fall in interest rates. In fact, only 30 to 40 percent of the firms reported that
they would expand investment in response to a fall in interest rates. One possible explanation
is that for firms, the indirect effect of lower interest rates, channeled through higher
aggregate demand, would provide a stronger incentive to expand investment than the direct
effect that results from lowering the cost of capital.

• Currency volatility was also perceived as an important constraint by firm CEOs. From
interviews with senior managers, we found that the response of firms to the currency
depreciation in 1998 was modest. This muted export response is puzzling especially because
there were many prevailing factors that were conducive to a marked increase in exports.
These factors included several government export promotion programs, excess capacity in 80
percent of the firms, forward contracts that permitted purchase of imports at a stable Rand
price, and strong global demand. Survey data do not address why firms did not export more
in response to a depreciated currency in 1998. Given that the survey was conducted in a year
when the economy was in near recession, it is possible that the incentive of a depreciated
Rand was not enough to compensate for the overall risk perception of CEOs or that concerns
over future volatility outweighed the benefits of depreciation.

      In conclusion, the main contribution of this survey is to provide an empirical basis for
identifying and ranking constraints confronting large manufacturing firms in GJMA and
South Africa. In this report, we confirm that crime and shortage of skills are binding
constraints to large manufacturing firm growth and that these areas need closer attention.
Moreover, our analysis shows that a piecemeal policy approach that resolves any one
constraint may not elicit desired the investment response from the firms. Instead, a
comprehensive approach aimed at attenuating all the constraints identified by firms will lead
to a more favorable environment in which the large manufacturing firms operate.
                                                    -1-




                                     1. INTRODUCTION

Motivation and objectives
In South Africa, nearly seven years into post-apartheid transition, concerns over low growth
and unemployment are widespread (see Box 1.1). The South African economy has undergone
structural change as it has become more integrated with global markets. Liberalization of
exchange rates and capital account also facilitated portfolio capital movements, leaving the
exchange and interest rates susceptible to macroeconomic shocks and capital volatility. Trade
liberalization exposed long-protected South African firms to foreign competition but in turn,
gave them access to larger global export markets. On the domestic front, a variety of labor
market reforms have been undertaken to rectify discriminatory practices of the apartheid era.
These reforms introduced fairer rules in the workplace but may have had unintended
consequences that hindered job creation. GDP growth rates slipped, giving rise to a persistent
low growth and job shedding throughout the 1990s. 3 Given South Africa’s history, supply-
side constraints have emerged as important impediments to growth. Fiscal discipline, low
external deficits and low inflation, which the government has succeeded in maintaining, are
indeed necessary for growth to occur. However, the enduring "low growth- low employment"
impasse in South Africa provides a compelling reason to disentangle the structural
constraints to growth.

       Fiscal decentralization places increasing responsibility on local governments to
contribute to national targets of growth and employment. To meet this responsibility, the
Greater Johannesburg Metropolitan Council (GJMC) adopted an innovative approach to city
development. They envisioned an economic information system to monitor the city’s
economy and to form the basis for policy interventions aimed at promoting local economic
growth and job creation. It led to a GJMC-World Bank partnership on the theme of local
economic development (LED). 4 The LED study involves surveying economic activity in
manufacturing and service sectors in Greater Johannesburg Metropolitan Area (GJMA). In
this report, the first in a series, we present the main findings of the large manufacturing firm
(firms with over 50 full time employees) survey in GJMA. 5

      Against the backdrop of structural changes in the South African economy, our primary
objective was to examine factors that constrain firm growth and employment in GJMA and in
the national urban economy. Based on a comprehensive firm- level database generated in
1999, this survey also aims at identifying key areas where efforts can be made to improve the
business environment in which these firms operate. To this end, we present what CEOs of
large manufacturing firms perceive to be the main constraints to growth and investment.
With the easing of international capital flows in recent years, CEO perceptions play a critical

3
  For a more through examination of growth performance and prospects in South Africa, see Lewis (2001),
“Policies to promote growth and employment in South Africa,” World Bank.
4
  A detailed methodology was developed for the LED study and is presented in Annex 1.
5
  We also surveyed firms in the large services sector, small, medium and micro enterprises (SMMEs), informal
sector firms, informal households, training and credit providers and production-level workers in GJMA. The
issue of non-income poverty monitoring is addressed separately.
                                                           -2-


       role in investment decisions and in understanding what deters investors from South Africa.
       We also describe, using follow- up interviews with senior managers, how these constraints
       affected the performance and operations of large manufacturing firms in 1998/99.

Box 1.1 Macroeconomic background

South Africa's post-apartheid government faced the challenge of promoting growth and employment while ensuring the
country’s transition in an environment of macroeconomic stability. Seven years later, the government's economic
policies have achieved mixed results. Table 1.1 summarizes key economic trends. Declining fiscal deficits, low and
falling external deficits, and the lowest inflation in decades can be counted among the macroeconomic success stories.
In the last several years, the government has taken steps to address other macroeconomic concerns. In particular, bank
lending rates have been brought down to a current level of 14.5 percent (they averaged 20 percent from 1996-98).

Table 1.1 Key economic indicators
                                               1994      1995       1996       1997       1998       1999       2000
GDP [growth, percent]                            3.2      3.1         4.2        2.5        0.7        1.9       3.1
Non-agricultural private sector employment      -0.9      0.5        -2.6       -2.5       -4.8       -1.2      -2.7*
[growth, percent]
Exports, GNFS [percent of GDP]                  22.2      23.0       24.5       24.6       25.9       25.9      29.1
Merchandise exports                             14.5      15.8       16.8       17.3       18.4       18.9      21.9
[net of gold, percent of GDP]
Merchandise exports                             19.6      23.9       20.0       13.6       14.6       11.4      27.1
[net of gold, growth in current Rand mls.]
Gross fixed capital formation by private        11.1      11.6       11.7       11.8       11.3       10.5      10.7
business enterprises [percent of GDP]
CPI [growth, percent]                           9.0        8.6        7.4        8.6        6.9        5.2       5.3
Interest rate [lending rate]                    15.6      17.9       19.5       20.0       21.8       18.0      14.5
Exchange rate [R/$, end-of-period]              3.5        3.6        4.7        4.9        5.9        6.2       7.6
Current account balance [percent of GDP]        0.1       -1.5       -1.3       -1.5       -1.8       -0.4      -0.3
* First three quarters
Source: South African Reserve Bank


Despite various Government initiatives, economic growth has not ensued. Of concern is the lackluster growth
(averaging 2.7 percent during 1994-00) which, combined with tight fiscal and monetary policies, has resulted in job
losses. While overall employment in the non-agricultural sectors declined by 6 percent, the private sector witnessed a
15 percent reduction in jobs during 1994-00. Gross fixed capital formation by private business enterprises fell by 3.6
percent in 1999. The sharp Rand depreciation in 1996 and again in 1998 did not result in sustained export growth.
Formal sector jobs continue to be shed, especially in manufacturing and mining.




       Sample

       Large manufacturing firms (with more than 50 full- time employees) were stratified on the
       basis of sector and size and a nationally representative random sample was selected (see
       Annex 2). We surveyed 325 large manufacturing firms spanning eight sectors. Map 1.1
       shows the zones included in the sample. The large firm survey (LFS) used seven
       questionnaires that were completed using personal interviews. The first questionnaire was
       used to elicit CEO perceptions on the prevailing investment climate and future prospects.
       This was augmented by detailed thematic questionnaires administered to different line
       managers to see how the performance and operations of firms were affected by the
       constraints identified by CEOs.
                                                    -3-



Map 1.1 Industrial zones and large firm sample




          Results from this survey have implications at a national le vel and for other metros
in South Africa. 6 This is because GJMA and its surrounding areas account for the bulk of the
firms and employment in South Africa’s formal manufacturing sector. They accounted for
approximately 40 percent of South Africa’s 6,100 large manufacturing firms in 1999 and
their share in national, formal manufacturing employment was approximately 42 percent.
Therefore, using appropriate weights, the sample can be used to draw implications at the
national level.
Sample characteristics
(i) Size-class: For the purpose of our analysis, we divided the sample into three size-classes
based on the number of full- time employees in 1999. Large manufacturing firms were

6
 A few location-specific constraints, like crime and infrastructure, have stronger regional implications. See
Annex 2 for a detailed description of the sampling technique and for a discussion on national and regional
weights.
                                                          -4-


classified into three size-classes: size 1 (50-99 employees), size 2 (100-199 employees), and
size 3 (200-10,000 employees). The sample was made up of 42 percent of size 1 firms, 27
percent of size 2 firms, and 31 percent of size 3 firms. As shown in Chart 1.1, size 1 firms
employed 12 percent of the workforce and size 3 firms employed about 73 percent of the
workers.

Chart 1.1 Distribution of full-time manufacturing employment in GJMA, 1998

                                                                Size 1
                                                                 12%

                                                                             Size 2
                                         Size 3                               15%
                                          73%




(ii) Sector: The LFS spanned eight sectors: chemicals, electronics and electrical machinery,
food and beverages, iron and steel, metal and metal products, paper and furniture, textiles,
and vehicles. A detailed breakdown of large manufacturing firms by sector is shown in
Annex 3. The sectoral distribution of firms in GJMA is similar to the distribution of firms in
the national economy, with minor exceptions.

(iii) Age: Where appropriate, we analyzed responses by age of firm. About 67 percent of the
firms have operated for over 10 years in South Africa; 28 percent are over 30 years of age.
Only 16 percent of firms were started post-apartheid coming into operation after 1995. 7 Age
and size are positively related: the average age for size 1 firms is 18 years, for size 2 firms 22
years and for size 3 firms 30 years. The oldest firms in South Africa are in the following
sectors: chemicals, iron and steel, metal products, paper and furniture, and textiles.

Chart 1.2 Age of large manufacturing firms in South Africa, 1999


                More than 30 yrs.                                        Less than 5 yrs.
                      28%                                                     15%




                                                                                 6 - 10 yrs.
                                                                                    18%
                                                                11-20 yrs.
                        21- 30 yrs.                               25%
                           14%



(iv) Ownership by Previously Disadvantaged Individuals (PDI): Given various initiatives to
promote PDI entrepreneurs, we examined their participation in the large manufacturing firm
sector. We found very few firms (around 5 percent) with PDI ownership. Size 3 firms (16

7
    See Annex 3 for a detailed analysis of young firms.
                                                -5-


percent) had a larger participation of PDIs on their boards as compared to size 1 firms (5
percent). Around 55 percent of the firms used PDIs in production jobs.


Chart 1.3 PDI participation in large manufacturing firms, 1999



               On the Board

            Only Ownership

                  Marketing

           Financial Manager

    Human Resource Manager

                 Production

           General Manager


                               0   10      20         30        40        50          60
                                                                           Percent of firms




Organization of the report
Chapter 2 provides the framework around which the rest of the report is structured. We rank
constraints to inve stment and growth as perceived by the CEOs and preview the main
findings of the report. We study how firms were affected by each of the constraints
described in Chapter 2 using information obtained from interviews with senior managers.
Chapter 3 scrutinizes location specific constraints. Chapters 4-5 present information on
financial and trade related constraints respectively. Chapters 6-7 present constraints relating
to labor regulations and skill shortages. Chapter 8 presents findings on regulatory
constraints. Additional detailed information is presented in annexes 1-9.
                                                        -6-


                   2. CONSTRAINTS TO BUSINESS GROWTH
In this chapter, we present CEO perceptions of factors constraining firm growth and lay the
groundwork for the rest of the report. More specifically, we rank constraints to investment
and growth as perceived by CEOs. Then, using information gathered from interviews with
senior managers, we examined the impact of constraints on firm performance and operations
in 1998/99. Although not strictly comparable, we present relevant findings from other cross-
country surveys to see if constraints to business in South Africa differ from those in
comparable countries.

      CEOs were asked to rate each constraint as a “major problem,” “moderate problem,”
“not a problem,” or “not applicable.” Using their responses, we constructed an index8 that
permitted a ranking of constraints. Chart 2.1 illustrates an index of CEOs rankings of the key
obstacles9 to business growth.

Chart 2.1 Index of CEO rankings on constraints to growth


    Environmental regulations

                 Infrastructure
         Business regulations

             Trade regulations

               Tax regulations
      Changing govt. Policies

                      Tax rates
    Availability of labour skills

    Corruption in government
    Recent labour regulations

         Depreciation of Rand
         Cost of capital/credit

               Crime and theft

                                    0   10   20   30      40      50      60      70     80      90     100
                                                                                                       Index




8
  To rank constraints, we created a weighted index using the ratings assigned by the CEOs to each constraint. If
a firm reported a constraint to be a major problem, the answer was weighted by 2. If a firm reported a
constraint to be a moderate problem, the answer was weighted by 1. Constraints that were not a problem or not
applicable were weighted by zero. Since the weights assigned were subjective, we tested the ranking by using
alternative weights and the current weighting system proved robust. The ranking reflects both the percent of
firms responding to the question and the severity of rating assigned to it. The rankings were statistically tested.
Annex 4 presents non-indexed CEO ratings of constraints by firm size.
9
  All results are significant at the 5 percent level unless otherwise stated. As a corollary, we also asked the firms
to rank what they would like the national and local governments to do for them. For the most part, the rankings
of what CEOs wanted from the local and national governments mirrored their rankings of constraints.
                                                    -7-


1. Our analysis of data shows that CEOs unambiguously ranked crime and theft as the
number one (assigned a value of 100) obstacle to firm growth. Around 95 percent of CEOs
perceive crime as a major to moderate problem. Follow- up interviews with senior managers
indicated that around 83 percent of the firms reported being victims of several types of crime
in 1998 (mostly employee theft and break-ins). Economic costs of crime were measured
using expenditures incurred by firms on crime prevention. On average, we found that firms
spent 1.6 percent of their sales revenue in 1998 on security guards and devices. Firms spent
between R 700 to R 2,500 per employee per year on crime prevention. We found a positive
and significant correlation between crime prevention expenditures and sales per employee
and investment. Employment expansion was negatively correlated with expenditure on
crime prevention. Of the firms surveyed, 61 percent mentioned that their employees were
victimized on their way to and from work. We found no discernible pattern in the spatial
distribution of crime. Details are presented in Chapter 3.

2. Firms ranked equally the cost of capital or credit, depreciation of the Rand, and recent
labor regulations as the next largest obstacle to firm growth. We study these rankings of CEO
perceptions in detail in Chapters 4-6.

      Cost of credit: Although 61 percent reported being adversely affected by interest rates,
responses of senior managers did not suggest that significant increase in investment will take
place in response to a fall in interest rates (Chapter 4). In fact, only 30 to 40 percent of the
firms reported that they would expand investment in response to a fall in interest rates. One
possible explanation is that for firms, the indirect effect of lower interest rates, channeled
through higher aggregate demand, would provide a stronger incentive to expand investment
than the direct effect that results from lowering the cost of capital.

      One-half of the firms surveyed used bank capital to finance their investment. Firms
mainly relied on internal savings to finance their working and investment capital. A higher
percentage of size 1 firms used bank financing to meet their investment needs as compared to
sizes 2 and 3 firms that relied heavily on internal savings. Therefore, size 1 firms were more
likely to be affected by the interest rate hike in 1998. In fact, survey results show that size 1
firms changed their structure of borrowing and investment plans as a result of the 1998
interest rate hike as compared to sizes 2 and 3 firms.

      Exchange rate movements: Depreciation of Rand was perceived as an important
constraint by firms CEOs. From other parts of the survey, we found that the response of firms
to currency depreciation in 1998 was modest (Chapter 5). This muted export response to
depreciation is puzzling especially because there were many prevailing factors that were
conducive to an increase in exports. These factors included several government export
promotion programs, 10 excess capacity in 80 percent of the firms, forward contracts that
permitted purchase of imports at a stable Rand price, and strong global demand. This survey
is unable to address why firms did not export more. Given that the survey was conducted in
year when the economy was in near recession, it is possible that the incentive of a
depreciated Rand was not enough to compensate for the overall risk perception of CEOs or
that concerns over future volatility outweighed the benefits of depreciation.

10
  Survey results also show that very few firms are aware of the DTI export promotion programs and even fewer
use them.
                                              -8-



       Recent labor regulations: CEOs ranked the recent labor regulations as an important
constraint to firm growth. However, around 70 percent of Human Resource Managers point
out that individual labor regulations had no effect on their employment decisions (Chapter 6).
Given that firms may be unable to pinpoint any individual regulation as being problematic,
we also asked them about their cumulative response to all four recent labor regulations
combined. Approximately 40 percent of firms reported that they undertook one or more of
the following in response to recent labor regulations: (i) hired fewer workers, (ii) substituted
labor with capital, and (iii) hired more temporary workers. However, 60 percent of the firms
still reported that all four labor regulations combined had no cumulative impact on their
employment decisions.

      We investigated potential unintended consequences of recent labor regulations that
might have contributed to raising the implicit cost of doing business with labor. In this
regard, we looked at the number of unions that the firms have to work with, the level at
which collective agreements regarding wages and benefits are struck, the number of strikes,
the resulting loss in workdays, the number of disciplinary actions taken against employees,
and the time and costs of hiring and firing workers. All these factors contribute to lowering
flexibility in the labor market. Firm responses were inconclusive.

      We found that the average number of unions that firms work with varies by size-class,
with 56 percent of size 3 firms dealing with 2 or more unions. Survey data also shows that
there is a positive relationship between firm size and level of collective agreements. On
average, 26 percent of firms strike agreements at the company level and as many as 40-50
percent of the firms reach agreements at the sector/industry level. This suggests reduced
flexibility for firms to decide on wages and benefits to reflect prevailing market conditions.
Firms did not experience many strikes, with 58 percent of all firms reporting no strikes over
the previous 12 months and 30 percent reporting one strike. Of the 42 percent of firms that
experienced strikes, approximately 45 percent lost 1 to 10 days and a third lost between 11 to
25 days. Around 62 percent of firms reported less than 1 disciplinary action for every 10
workers and retrenchment costs (in months taken to retrench) were around 0-2 months for
two-thirds of firms surveyed.

      In keeping with global trends, we found that many South African firms rely on
subcontracting and using temporary workers. Subcontracting and use of temporary workers
varied directly with firm size. Approximately 80 percent of the firms noted that using
temporary labor gave them the flexibility to respond to market conditions. Around 50
percent of firms reported that they subcontracted work in order to respond to surges in
workload and to find specific skills.

3. Corruption in government was rated as the next highest obstacle to firm growth. We
focused on this issue by exploring whether the regulations and procedures for obtaining
government licenses and permits were cumbersome or corrupt. Follow-up interviews with
senior managers suggest that corruption in government did not appear to be as problematic as
its ranking suggests (Chapter 8). Approximately 1 percent of firms reported that they were
asked to pay bribes (although the perceived “official” nature of our survey may have affected
these responses).
                                              -9-


4. CEOs ranked the availability of labor skills and level of taxes as the next constraint to
firm growth. LFS evidence on tax rates does not substantiate the importance accorded to
them in the CEOs rankings while lack of skilled labor appears to be a far more binding
constraint than its ranking suggests (Chapter 7).

      In fact, approximately 80 percent of firms reported extreme to moderate difficulty in
finding people with managerial and professional skills. Around 70 percent of firms reported
extreme to moderate difficulty in finding people with service and craft skills. Given South
Africa’s apartheid past, we found that skills were closely tied to race, with whites dominating
the managerial/technical positions and Africans occupying the unskilled positions. If finding
people with appropriate skills was difficult in a year when the economy was in near
recession, it is likely that skill shortages could become a binding constraint when the
economy starts to grow at a rapid pace.

      Moreover, despite the fact that firms found it difficult to find skilled workers, we found
that only 35 to 45 percent of firms surveyed provide training to approximately a third of their
workers. Firms were reluctant to reveal how much they spent on training. We found no
compelling evidence to show that firms were reluctant to invest in training because they were
apprehensive about losing their trained workers to other firms. Less than 5 percent of trained
workers left the firms after being trained.

5. Finally, tax regulations, business regulations, and trade regulations were also ranked as
important constraints to firm growth. Factual evidence shows that tax and business
regulations (Chapter 8) do not appear to be a problem. Firm responses to trade regulations
(Chapter 5) are discussed below.

       We examined the effect of tariff and non-tariff regulations on large manufacturing
firms. Due to lower tariffs in the 1990s, approximately 60 percent of size 1 firms (compared
to 46 percent of size 3 firms) noted experiencing a reduction in product prices. Compounding
the problem, fewer size 1 firms reported benefiting from a fall in prices of imported inputs
due to tariff reductions. We also found that 30-40 percent of firms reported being adversely
affected by dumping by foreign firms and by illegal customs control procedures. By sector,
approximately 85 percent of textile firms reported being adversely impacted by dumping and
by illegal customs procedures.

6. In contrast to crime, infrastructure, the other location-specific variable, was assigned a
relatively low rating of 25 suggesting that CEOs did not find it a major bottleneck.
Reiterating their positive ranking of infrastructure, around 85 percent of CEOs indicated that
GJMA was their preferred choice of location for expansion.

      Firms were asked to rate their current location on the basis of proximity to product and
input markets, availability of services and on reliability and costs of services (Chapter 3).
LFS data shows that most firms rated their location as excellent to fair with regard to
proximity to product and input markets. In general, availability of services also received a
positive rating from firms. With respect to costs and reliability of services, we found that
costs of services were consistently rated lower than their reliability. International and
domestic telecommunication service received relatively poor ratings on cost and reliability,
                                                                -10-


with 35 percent of the firms reporting the costs as high. Again, 70-75 percent of the firms
gave their location a poor rating with respect to crime.

International comparisons

To put the findings from our survey into an int ernational perspective, we present results from
the Asian Manufacturing Recovery Survey of 1998 in Table 2.1. This survey was conducted
to take stock of the middle- income Asian economies after the Asian financial crisis in 1997.
Because the lists of constraints differed, the ratings presented in Table 2.1 are not directly
comparable to those in Chart 2.1. Nevertheless, the table provides a useful point of reference.

       Although crime and exchange rates are not options in Table 2.1, the CEOs in the 5 East
Asian economies rated high interest rates as the number one problem. Other highly ranked
obstacles were the supply of skilled labor (Thailand and Malaysia), and corruption in
government (Thailand and Philippines). Labor regulations were ranked relatively low in
Table 2.1, but labor costs were ranked relatively high in East Asia. 11 Access to finance and
red tape were identified as key constraints in East Asia. East Asian firms also enjoy
relatively good infrastructure and a relatively hassle free business and trade regulatory
environment.


Table 2.1 Rankings of constraints to business growth in 5 East Asian countries*
                                      Thailand         Philippines           Indonesia             Korea            Malaysia
High interest rates                                          1                   1                   1                 1
Labor costs                                1                 2                   2                   4                 4
Red Tape                                   2                 3                   6                   2                 6
Access to finance                          4                 4                   4                   3                 3
Supply of skilled workers                  3                10                   5                   5                 2
Quality of suppliers                       6                 6                   3                   8                 5
Corruption                                 5                 5                  10                   6                11
Customs administration                     8                 7                   9                   7                 9
Duty exemptions/tarriffs                   7                 8                  11                   9                 7
Infrastructure                             9                 9                   7                  11                10
Labor regulations/unrest                                    11                   8                  10                 8
*Simplified from original report for comparability. Source: Asia Manufacturing Recovery: A firm level analysis, World Bank, 2000


      The World Business Environment Survey (WBES, 2000) provides a comparison of the
business environment in South Africa with Latin American and OECD countries (Table 2.2).
Except for Mexico, Thailand (not shown here) and South Africa, crime is not a serious
problem in the other countries. In the case of exchange rates, South Africa does better than
Mexico, falls short of European countries and the U.S., and can be bracketed with other Latin
American countries and the U.K. In finance, South Africa is comparable to OECD and ranks
well above Mexico. However, in labor regulations, South Africa falls short of OECD
countries (except Italy), falls even below Mexico, and is comparable to Brazil and Argentina.




11
     In our survey, firms were not willing to comment on labor costs directly.
                                                      -11-


Table 2.2: Constraints to business growth in Latin America, OECD, and South Africa*
                     South    Argentina   Brazil   Chile   Mexico   UK    Germany   Spain      Italy   United
                     Africa                                                                            States
Crime                  3.6       2.5       2.7      2.5      3.4    2.0     1.7       1.9       2.1     2.2
Finance                2.4       2.9       2.7      2.4      3.2    2.3     2.5       2.2       2.1     2.3
Exchange Rate          2.4       1.8       2.9      2.5      3.2    2.3     1.7       1.9       1.9     1.6
Policy Instability     2.0       3.1       3.5      2.6      3.3    2.2     1.6       2.4       2.9     2.0
Inflation              2.5       2.0       2.7      2.2      3.4    2.2     1.9       2.3       2.2     2.2
Corruption             2.6       2.6       2.5      1.9      3.3    1.3     1.8       2.1       1.8     1.8
Labor                  3.2       3.0       3.5      2.5      2.8    2.6     2.8       2.4       3.0     2.3
Regulations
Taxes                  2.8       3.3       3.6      2.2      3.2    2.8     3.1       2.6       3.3     2.4
Business               2.0       2.4       2.7      2.3      3.2    1.9     2.2       2.2       2.6     2.1
Regulations
Infra-structure     1.8         1.9        2.2      1.9     2.3     1.6      1.8      1.9       2.3     1.8
AntiCompetitive     na          2.4        2.4      1.9     2.8     1.7      2.2      2.3       2.2     1.7
  *A score of 1 means no obstacle and a score of 4 means major obstacle. Source: WBES, 2000.

        For most macroeconomic variables, including infrastructure and financial services,
  South Africa enjoys a rating closer to OECD than Latin America or East Asia. In many of
  the East Asian countries, firms are relatively more constrained by red tape and business
  regulations, access to financing and labor costs. In addition to these constraints, firms in
  Latin America face corruption and policy instability. For South African firms, the business
  environment is relatively free of these constraints. However, in the case of crime and labor
  regulations, South Africa’s rating falls below OECD and comparable countries in Tables 2.1
  and 2.2.
                                                                     -12-


                                           3. LOCATION CONSTRAINTS
Our analysis of CEO responses shows that GJMA received a mixed ranking as an industrial
location. On the one hand, crime is ranked as the main constraint to firm growth in GJMA;
but on the other hand, GJMA’s industrial location received a positive ranking on
infrastructure and service delivery. Both crime and infrastructure are location specific;
therefore the discussion in this chapter applies only to GJMA and its surrounding areas.

Crime and violence in GJMA
Crime and violence were rated as the number one constraints to business growth by an
overwhelming 95 percent of CEOs. Of these, 70 percent described it as a “major” obstacle to
growth, and another 25 percent labeled it a “moderate” obstacle. Only 5 percent said that
crime was “not a problem.” When asked what they would like the local government do for
them, 62 percent of the firms mentioned that safety and security was the top priority. These
statistics reveal that crime is a major problem and has economic costs associated with it.

Chart 3.1 Types of crime in GJMA, 1998


                     70
                     60
                     50
  Percent of firms




                     40
                     30
                     20
                     10
                      0
                          Employee Break-ins &   Other   Vandalism    Extortion   Arson   No crime
                            theft    property
                                       theft

      Approximately 83 percent of the firms reported being victims of several types of crime
in 1998. Employee theft, break- ins and property theft were the most prevalent forms of
crime experienced by about 60 percent firms (Chart 3.1). Of the firms surveyed, 61 percent
reported that their employees were victimized on their way to and from work. Only 16
percent of the firms were free of crime and violence in 1998. We found no pattern in the
spatial distribution of crime.


Costs of crime prevention in GJMA

We use expenditure incurred by firms on crime prevention (on security guards and devices)
as a proxy for the economic costs of crime. In 1998, we found that expenditures on crime
prevention averaged about 1.6 percent of sales revenue. Evidence also indicates that these
expenditures increased between 1997 and 1998, which might be indicative of worsening law
and order conditions. For about 49 percent of firms, the average expenditure increase was
approximately R 210 per employee per year.
                                                      -13-


Table 3.1 Firm spending on crime prevention, 1998 (Percent of firms)

         Rands per employee                           Size 1            Size 2          Size 3
Less than 700                                          55.4              41.5            35.5
701-1000                                               12.1              18.4            12.6
1001-2500                                              20.7              25.7            44.6
More than 2500                                         11.7              14.4            7.31

      The firms surveyed spent on average R 700 (over US $ 100 at the 1998 exchange rate)
to over R 2,500 per employee per year in 1998. As shown in Table 3.1, expenditure on
guards and security devices varies positively with firm size. 12 For example, 55 percent of size
1 compared to only 35 percent of size 3 firms spent less than 700 Rands per employee per
year. Compared to only 21 percent for size 1 firms, about 45 percent of size 3 firms spent R
1001- 2500 (US $ 151–400) per employee per year. Approximately 7-14 percent of all firms
spent more than R 2,500 per employee per annum in 1998.

Relationship between crime and economic variables

If we assume that expenditure on crime prevention is an indicator of the intensity of crime
that firms are subject to, survey data can be analyzed to examine the relationship between
crime prevention expenditures and other economic variables such as sales, investment,
employment, growth, and age.

Table 3.2 Correlation of expenditure on crime prevention with economic variables

                                            Correlation coefficient        P > |R| under Ho: Rho = 0
Sales per employee in 1998                         0.36323*                          0.0001
Investment                                         0.17431*                          0.0048
Employment98/ Employment97                        -0.18759*                          0.0030
Investment expansion                               -0.01977                          0.7773
Age                                                -0.03292                          0.5886

      Table 3.2 indicates a significant and positive but low correlation between annual crime
prevention expenditures and sales per employee and investment. This suggests that firms
that were performing better in 1998 also spent more money on security guards and security
devices. Employment expansion was negatively correlated to crime prevention expenditure.
Infrastructure and service delivery in GJMA
Good infrastructure and efficient service delivery contribute positively to business growth by
influencing a potential investor’s decision to establish a firm in a particular location,
reinforcing an existing firm’s decision to expand in its present location, or preventing it from
shifting operations to another location. The LFS data indicates that the CEOs give a positive
rating to the availability and quality of infrastructure and services in GJMA. 13

12
     The results are significant at the 10 percent level.
13
     This finding is corroborated by the World Business Environment Survey, 2000 as discussed in Chapter 2.
                                                                -14-



Rating of present Location

Firms were asked to rate their present location in terms of proximity to product and input
markets. Chart 3.2 shows that 95 percent of the firm managers assigned an excellent or fair
rating to “proximity to main road.” Between 70-88 percent of firms assigned similar ratings
to other proximity related variables such as international airport facilities, product and input
markets, suppliers of machines and parts, and corporate financial services. No more than 6
percent of the firms assigned a negative rating to these variables. From its inland location,
GJMA’s firms export mainly to other SADC markets. A relatively smaller percent of firms
(around 60 percent) rated proximity to SADC export markets, rail transport and proximity to
semi-processed inputs 14 as excellent to fair. Locations in GJMA that are not well connected
by rail transport are marked on Map A5.1 in Annex 5. Except for proximity to seaport, none
of the other proximity-related factors received a negative rating from more than 10 percent of
the firms.

Chart 3.2 Firm ratings of location on proximity to infrastructure and services, 1999


                                            Excellent            Fair             Poor


                     Prox. to seaport
     Prox. to SADC export markets
          Prox. to semi-proc.inputs
                Prox.to rail transport
               Prox.to raw materials
 Prox. to suppliers of mach./parts
           Prox. to corp.fin.services
       Prox. to natl. airport facilities
           Prox. to product markets
        Prox. to intl.airport facilities
            Prox. to main road links

                                           -50      -30   -10           10   30          50    70         90
                                                                                              Percent of firms




      Chart 3.3 shows that firm managers rate the availability of most services as excellent or
fair with the availability of courier services rated the highest by about 96 percent of firms.
The two services that are viewed as being relatively less available are industrial land and
waste facilities. Yet even in the case of these, no more than 11-17 percent of firms assigned
them a poor rating.



14
     All results are significant at the 5 percent level unless otherwise stated.
                                                                              -15-




Chart 3.3 Firm ratings of availability of services, 1999


                                                                              Excellent          Fair           Poor


   Availability of industrial waste facilities

               Availability of industrial land

         Availability of container services

           Availability of trucking services

            Availability of courier services

                                                      -20             0              20          40         60           80           100
                                                                                                                          Percent of firms


       Firms were asked to rate reliability and cost of services. Chart 3.4 indicates that
reliability of water and electric services scored the highest rating from 90 percent of firms.
Costs of services were consistently rated lower than their reliability. International and
domestic telecommunication services scored relatively poor ratings for cost and reliability.
Around 35 percent of firms rate telecommunication costs as high. Again, neighborhoods in
which firms rated telecom poorly are graphed on Map A5.2 in Annex 5. The most important
point to glean from Chart 3.4 is the extremely poor rating firms gave to the crime rates
affecting both their staff and their company property. Close to 75 percent of firms noted
their staff was affected by crime, while another 70 percent observed that crime affected their
property.

Chart 3.4 Firm ratings of location on cost and availability of services, 1999
                                                                  Excellent         Fair       Poor


     Low crime rate affecting firm property
             Low crime rate affecting staff
                        Low local tax rates
         Low cost of intl telecom services
       Low cost of dom. telecom services
                     Low cost of electricity
                          Low cost of water
       Reliability of dom. telecom services
                Low cost of industrial land
         Reliability of intl telecom services
                  Low cost of office space
               Reliability of electric supply
                 Reliability of water supply

                                                -80         -60       -40        -20       0          20   40      60       80        100
                                                                                                                        Percent of firms

      In general, the LFS reveals that most large manufacturing firms in GJMA found
infrastructure and service delivery as fair to excellent. Key areas needing improvements are
                                                   -16-


crime prevention, improvement in telecommunication services, industrial land and industrial
waste facilities, and improvement in access to rail transport in select locations to facilitate
proximity to SADC markets.
Choice of Location for Expansion
Firm managers were asked where they would choose to locate if they were to expand their
production. As seen in Chart 3.5, a majority of firms favored their present location.
Approximately 60 percent of the firm managers in GJMA would prefer to stay on the same
premises. Around 85 percent of firm managers noted that they would expand within GJMA.
Only 12 percent of firm managers said they would expand outside of GJMA, with 3 percent
of that group preferring to expand entirely outside South Africa. The main disadvantages of
their present location were high crime and violence, long distance to a seaport and pollution.
Chart 3.5 Choice of location for expansion (percent of firms)


                                 Outside Gauteng
           Outside Jo'burg                         Outside South Africa
                                       4%                  3%
                 2%                                              Not Applicable
                                                                        3%
    Industrial area outside
            Jo'burg
               3%

         Other area in Jo'burg
                 13%

                                                                              Same premises
                                                                                  60%
                          Same industrial area
                                 12%




      Overall, GJMA rates high on provision of infrastructure and services but crime remains
a major deterrent. Crime notwithstanding, a large majority of firms indicated that they prefer
to expand within the GJMA area.
                                                     -17-


                               4. CORPORATE FINANCE
Gross fixed capital formation by private business enterprises in South Africa fell from 8
percent during 1996-97 to –3.6 percent in 1999. During this period, real lending rates rose
from around 6.6 percent in 1994 to 15 percent in 1998 and fell to approximately 11 percent
by the end of 1999. This chapter evaluates the effect of the 1998 interest rate hike on
investment decisions during 1998-99. Before examining how firms adjusted due to the
interest rate hike, we looked at the sources of investment capital.
Sources of financing
Survey evidence suggests that access to bank loans is not an issue in the case of the large
manufacturing sector firms. This is probably because a majority of the large manufacturing
firms have been operating since the 1970s and have well-established links with financial
institutions. Despite access to bank capital, the LFS records that only about 50 percent 15 of
firms used bank loans to finance their operations in 1998-99. Sources of investment capital
reflect the following. First, it appears that firms preferred internal savings to other sources of
available financing. Second, it is likely that during 1994-98, in response to high and
oscillating real interest rates, LFS firms were reducing their reliance on bank financing even
before the 1998 interest rate hike.

        For most firms in GJMA, the main source of financing was two forms of internal
savings: retained earnings and loans from a parent establishment at subsidized interest
rates. 16 Chart 4.1 shows that a larger percentage of size 3 and size 2 firms relied on internal
savings to finance their investment needs as compared to size 1 firms. We also found that a
higher percent of size 1 firms (60 percent) relied on local bank financing as compared to size
2 and 3 firms (40 percent). In fact, data showed that 32 percent of firms relied on bank
capital to finance as much as 50-100 percent of their investment needs. 17 As discussed in the
next section, since size 1 firms used more bank capital, they were likely to be more adversely
affected by the interest rate hike in 1998.




15
   Higher interest rates must have also indirectly affected those 50 percent of the LFS firms that did not use
bank loans through an increase in the opportunity cost of capital associated with internal savings.
16
   Retained earnings were combined with loans from the parent establishment into a grouping called “internal”
sources of financing due to the negligible interest rates charged by parent establishments.
17
   The sources of financing of working capital were similar to those of investment capital in 1998 and were also
financed mostly from internal savings.
                                                               -18-


Chart 4.1 Sources of investment capital by size -class, 1998-99
                                                                Size 1


                                   Other

    Shares issued on the stock exchange

       Loan from partner or parent estab.

                 Loan from a foreign bank

    Loan from a local South African bank

              Retained/internal earnings

                                            0        10        20        30       40          50        60        70
                                                                       Percent of firms




                                                                Size 2


                                   Other

    Shares issued on the stock exchange
       Loan from partner or parent estab.

                 Loan from a foreign bank

    Loan from a local South African bank

              Retained/internal earnings

                                            0        10        20        30      40        50           60        70
                                                                      Percent of firms


                                                                      Size 3


                                        Other

        Shares issued on the stock exchange

           Loan from partner or parent estab.
                     Loan from a foreign bank

        Loan from a local South African bank

                    Retained/internal earnings
         1-50%
                                                 0        10    20       30      40       50       60        70        80
         51-99%
                                                                           Percent of firms
         100%
                                                                              -19-


Firms’ responses to the interest rate hike
LFS data indicates that about 61 percent 18 of firms reported a change in investment or
borrowing behavior as a consequence of the 1998- interest rate hike. An additional 11
percent of firms originally reported the response “other” that, upon closer examination,
indicates that these firms did not change investment behavior but simply experienced reduced
profits and/or ability to extend credit to customers. 19

Chart 4.2 Firm responses to interest rate hike, 1998
                                                                Size 1         Size 2           Size 3

                        45
                        40
                        35
     Percent of firms




                        30
                        25
                        20
                        15
                        10
                         5
                         0
                             Change term structure   Reduce level of     Reduce planned plant            Not at all   NA
                             of borrowing and new      borrowing             expansion
                                    liabilities


      Chart 4.2 summarizes the key responses of firms to high interest rates. About 40
percent of size 1 firms reduced planned investment and/or reduced their level of borrowing
and about 30 percent of them changed the term structure of their borrowing. Only 22 percent
of them reported no response to the 1998- interest rate hike (“not at all” plus “NA”). About
30 percent of size 3 firms reported reducing planned plant expansion and their level of
borrowing, while a slightly smaller proportion (26 percent) changed their debt-equity
structure. In contrast to the size 1 firms, about 40 percent of size 3 firms reported “no
response” to the 1998 interest rate hike. Therefore, size 1 firms were more likely to change
investment plans as compared to size 3 firms because of the interest rate hike. These results
are consistent with the fact that size 1 firms use relatively more bank capital for investment
and were directly affected by the high cost of capital.


      LFS data indicates that 30 to 40 percent of the firms are likely to respond by increasing
investment to the drop in the interest rate from 15 percent in real terms in 1998 to 13 percent
in 1999. How can these numbers be reconciled with 61 percent of the firms who noted that
they had been affected negatively by the interest rate hike and with 80 percent of CEOs who
noted that high interest rates were the second most important obstacle to firm growth? One
possible answer is that for these firms, the indirect effect of lower interest rates channeled

18
   While less than 50 percent of the firms marked any one response in Chart 4.2, it is important to note that it
was possible for a firm to respond to one or all three actions. Hence the cumulative response from 61 percent of
all firms which noted that they were affected negatively by the interest rate crisis.
19
    11 percent of firms that reported “other “ and are reclassified are not shown in Chart 4.2. Upon further
investigation, 20 percent of these firms reported being affected by the interest rate hike through higher costs and
lower aggregate demand.
                                              -20-


through higher aggregate demand is a stronger incentive for investment than the direct effect
that simply reduces the cost of capital. This suggests that while a significant decline in real
interest rates is considered necessary by CEOs, it is not a sufficient trigger for strong
investment growth and job creation in the large manufacturing tier of the South African
economy.

Other firm characteristics
In addition to the interest rate hike in 1998, firms also experienced a macroeconomic shock
in the form of a significant real depreciation of the exchange rate. Was the response to the
interest rate hike different among exporters and non-exporters? Statistical analysis reveals
that the exporter status of LFS firms did not display any significant differences in their
interest rate response.

      We also examined whether the legal status, age or number of years of operation, or
trading characteristics determines the investment pattern of firms. Statistically, no discernible
differences were found to validate any of these hypotheses. It appears that despite well
developed capital markets in South Africa, firms across all sectors, age, legal status, etc. rely
to a large extent on retained earnings and loans from parent establishments to finance both
investment and working capital.

     We also grouped firms on the basis of those that expanded by increasing investment in
1998-99 relative to those that did not. The cut-off point was an increase in investment in
excess of 10 percent of the investment-assets ratio. About 50 percent of the LFS expanded
according to this definition. Firms that increased investment between 1997 and 1998 are
shown in Map 4.1.

       We also looked at firms that expanded by increasing both investment (according to the
criteria defined above) and employment. The cut-off point used to differentiate one group
from the other was a minimum of 5 percent of employment expansion applied to the base
level of employment that distinguishes each size class. For example, given that size 3 covers
firms over 200 employees, the minimum increase in employment required for a size 3 firm to
qualify in the “expanded through investment and employment” grouping is 10 employees or
5 percent of the employment threshold. The equivalent number for size-class 2 is 5 and for
size-class 1 is 3 new employees. Using this definition, approximately 14.6 percent of the
firms were identified as “expanding through both investment and employment.” Map 4.2
shows the firms that increased both employment and investment.
                                       -21-


Map 4.1 Firms that increased investment between 1997 and 1998




Map 4.2 Firms that increased both investment and employment between 1997 and 1998
                                                     -22-


               5. EXCHANGE RATE AND TRADE POLICY
The reintegration of South Africa into global markets and process of trade liberalization led
to increased access to global markets but in turn, exposed South African firms to competition
in domestic and foreign markets. Moreover, transition to a unified and flexible exchange rate
regime and partial liberalization of the capital account introduced greater volatility into the
exchange rate- interest rate systems, and made the economy more susceptible to external
shocks. A case in point is the Asian contagion that prompted the Rand crisis of 1998. In this
chapter, we examine the responses of the firms to the exchange rate depreciation in 1998 and
to changes in tariff and non-tariff barriers that occurred in the 1990s. We also report the
extent to which firms are aware of and make use of the export promotion programs.

Trading profile of firms

As many as 70 percent of the large manufacturing firms engage in international trade. The
ability to trade is positively related to firm size rising from 56 percent for size 1 firms to 85
percent for size 3 firms. Although a large proportion of surveyed firms trade, they export
only 13-17 percent of their output. Even though a larger proportion of size 3 firms export, the
proportions of their output exported or imported are no larger than those of size 1 firms.
There is little variation across sectoral lines - the sectors that export most are vehicles and
parts (27 percent), iron and steel (23 percent), and electronics and electrical machinery (19
percent). 20 While a strict comparison is not possible and the scales of production need to be
examined, Box 5.1 suggests that compared to East Asian exporters, South African
manufacturers exported a significantly smaller proportion of their output in 1997-98.

           Box 5.1 Average proportion of manufacturing output exported in
           East Asia in 1997-98

                  Country                             Mean %               Median %
                  Indonesia                           36                   28
                  South Korea                         37                   27
                  Malaysia                            54                   50
                  Philippines                         67                   61
                  Thailand                            57                   66

                  Source: East Asia Surveys, World Bank, 1999.

      As Table 5.1 shows, SADC is the leading destination for 78 percent of the firms that
export approximately 66 percent of their total exports to this market. 21 The rest of Africa is
the second most-popular market for 32 percent of firms that export 26 percent of the total
exports to it. In third place and outside Africa, are the West European, Australasian and
Asian export markets, with Western Europe receiving 43 percent of total exports from GJMA
compared to Asia and Australasia that receive about 20 percent each. Although only 14

20
   The proportion of inputs imported is about 35 percent on average. Annex 6 provides more details along
sectoral lines.
21
   This pattern is likely to differ in other metropolitan areas with port facilities. Because of Johannesburg’s
location, a regional rather than a national perspective is more appropriate for this section.
                                                        -23-


percent of the firms export to North America, the proportion of their exports to this
destination is about 37 percent.
Table 5.1 Exports by destination

                                      Percent of firms exporting to Proportion of exports sold to
                                            this destination:        this destination (percent)
SADC                                               77.5                         65.6
Rest of Africa                                     32.4                         26.1
Western Europe                                     19.8                         43.3
Central/East Europe                                14.6                         27.9
Asia                                               18.6                         20.8
Australasia                                        20.6                         20.9
N. America                                         13.8                         37.4
Rest of Americas                                   14.9                         25.3
Note: Since firms export to more than one market, the numbers need not add up to 100.

Exchange rate shocks and export behavior

Firms responded to the Rand crisis of 1998 in three ways: 1) they exported more; 2) they
substituted away from imports, or 3) took no specific action. Chart 5.1 illustrates that among
the exporting firms, size 3 firms tended to export more (30 percent increased exports
significantly and about 50 percent moderately), while size 1 or 2 firms responded through
greater import substitution. This is consistent with the fact that size 3 firms are more export
oriented. Almost 30 percent of size 1 or 2 firms were uncertain about exchange rate
depreciation and decided to wait and see compared to less than 20 percent of size 3 firms. So
although the export response was positive, it was muted suggesting that South African firms
were unable to exploit exports as a channel of growth.

Chart 5.1 Firm response to depreciation, 1998

                                               Very Much   Moderate    Not at all

              Size 1
       Wait and see
     Exported more
  Import substitution
              Size 2
       Wait and see
     Exported more
  Import substitution
              Size 3
       Wait and see
     Exported more
  Import substitution

                        0   10   20       30       40          50     60        70        80       90   100

                                                                                     Percent of firms
                                                    -24-


    The inability of South African large manufacturing firms to export more, despite having
excess production capacity, is puzzling. This survey cannot directly explain why large
manufacturing firms did not export more despite Rand depreciation. A possible explanation
might lie in the weak export performance of the size 1 firms. As in the case of Mexican firms
(Box 5.2), performance of smaller firms might be more closely tied to domestic markets
rather than to global economic conditions and depreciation of the exchange rate. It is likely
that the incentive of a depreciated Rand was not sufficient to compens ate for the overall risk
perception of the CEOs.

Box 5.2 Export Dynamics and Productivity in Mexico

A study of manufactured exports in Mexico during the 1990s revealed that after trade liberalization, more firms
have begun exporting, but the most pronounced growth has been in large-scale and foreign owned firms. Export
orientation of smaller firms is determined more by short-term domestic demand conditions (i.e. during domestic
recession), while large firms have a more permanent presence in export markets. High-performance workplace
practices and labor skills are linked with exporting and productivity growth. Lack of access to financing is an
important constraint from smaller firms both as a limit to pre-entry and post-entry export expansion
The study finds:
• The pre-entry-to-trade phase is important. Building labor skills, investing in capital and establishing quality
     management systems is important. Governments should focus on extending support for enterprise training,
     technology diffusion and information.
• Buyer-supplier relationships extend productivity benefits of exporting. Supplier development programs to
     encourage large firms to buy from smaller firms/subcontractors should be developed and implemented in
     effective ways to facilitate private-to-private transactions.
• Productivity improves more when exporters have more experience. Governments need to understand
     transitory exporters.
Source: Export Dynamics and Productivity: Analysis of Mexican Manufacturing in the 1990s



Competitors in global markets
Table 5.2 illustrates the top 5 foreign competitors of LFS firms in the main global export
markets as noted by their sales/marketing managers. The numbers in brackets indicate the
actual number of the GJMA firms that compete with foreign firms. Table 9.1 indicates that in
the SADC market, GJMA’s large manufacturing exporters or LFS firms compete mostly with
other South African firms or firms from other SADC countries. In the rest of Africa, Kenya,
Uganda and the United States are the main competitors. Ugandan firms are the fourth largest
competitors and 22 (out of 325) GJMA exporters compete with them. In other regional
markets, especially Europe, Australia and North America, local firms provide the toughest
challenge to GJMA’s large manufacturers. In particular, in these markets, the absence of
competition from the East Asian countries and China is conspicuous.
                                                       -25-




Table 5.2 Top five foreign competitors in key regional export markets


Market              Competitor 1      Competitor 2        Competitor 3        Competitor 4      Competitor 5

SADC                 South Africa     Botswana (155)      Zimbabwe (97)        Namibia (81)     Swaziland (49)
                        (All)

Rest of              South Africa     Kenya     (30)          U.S.     (24)    Uganda (22)      Zimbabwe (17)
Africa                   (59)

Western            Germany (109)       U.K.    (101)          Other Europe     Italy    (51)    France      (42)
Europe                                                            (52)

Eastern             Germany (36)       U.K.     (35)          France   (24)   Bosnia & Herz.    Bulgaria (14)
Europe                                                                             (14)

Asia                Taiwan    (55)     India    (43)          South Korea      Thailand (28)    Indonesia (24)
                                                                  (42)

Australasia         Australia (110)    New Zealand        Philippines (15)     Papua New          Italy (11)
                                          (31)                                 Guinea (13)

North               U.S.     (128)     Spain     (9)          Europe    (9)   South Korea (1)   Japan        (1)
America

Rest of             Chile    (26)     Argentina (17)      Europe        (7)    Brazil    (1)            .
Americas
Note: Figures in brackets indicate the number of GJMA’s manufacturing firms that compete with any particular
country in that market.


Exchange rates and forward contracts
In the aftermath of the 1996 crisis, many large firms purchased forward contracts from the
South African Reserve Bank or commercial banks to protect themselves against exchange
rate shocks. This would help protect them from incurring higher costs of imported inputs
while being able to benefit from higher exports in the event of exchange rate depreciation.
When the Rand crisis of 1998 unfolded, firms’ trade responses were influenced by their
hedging status.

      LFS evidence indicates that hedging in 1998 was positively correlated with firm size.22
Chart 5.2 illustrates that about 25 percent of size 1, 50 percent of size 2 and 58 percent of
size 3 firms purchased forward contracts. On average, about 43 percent of all firms were
hedged. For all firms, if a forward contract was purchased, the average percent of the
currency exposure hedged was 95 - 100 percent. 23 At the sectoral level, the sectors in which

22
  This result is significant at the 5 percent level.
23
  Both size 1 and 3 firms tend to hedge 100 percent of currency exposure. The average currency exposure
hedged for size 2 firms was, however, only 50 percent.
                                                              -26-


the largest number of firms were hedged were electronics and electrical machinery (70
percent) followed by chemicals (55 percent) and vehicles (51 percent). The least number of
firms hedged was in textiles (20 percent).

Chart 5.2 Forward contracts on foreign currency exposure, 1998



                Vehicles
                 Textiles
          Paper/Furniture
          Metal Products
               Iron/Steel
         Food/Beverages
                Electrical
              Chemicals


                   Large
                 Medium
                   Small

                             0        10      20         30           40   50   60        70         80

                                                                                 Percent of firms

       Since hedging through forward contracts protects a firm against foreign exchange rate
risk, it is important to discern whether firms that were 100 percent hedged behaved
differently from those that were not hedged at all. 24 This section focuses on how hedged
firms changed their exporting or importing patterns.

Chart 5.3 Comparing response to currency de preciation, 1998 (All firms 100 percent
hedged and not hedged at all)


           Don't hedge Started
             exporting more
       Hedge Started exporting
               more
      Don't hedge -More import
             substitution
           Hedge -More import
              substitution

                                  0            20                40        60        80             100
        Very much                Moderately         Not at all                             Percent



     Chart 5.3 displays the two main responses of firms to the 1998 exchange rate
depreciation. They were “started exporting more” and “engaged in more import substitution.”
Almost 60 percent of the firms that were hedged 100 percent starting exporting more

24
     All results are significant at the 10 percent level.
                                                     -27-


compared to less than 40 percent of those that were not hedged. There was no significant
difference in import substitution among those hedged 100 percent and those not hedged.
About 35 percent of each group engaged in greater import substitution.

      An analysis of size 2 and 3 firm responses shows that the share of those who were 100
percent hedged and exported more was 67 percent compared to 46 percent for those who
were not hedged. In general, the forward contracts did not affect in any significant way either
the exporting or importing responses of size 1 firms. This is also validated by the fact that
only 25 percent of size 1 firms bought forward contracts (see Annex 6 for more details).

Trade policy
Tsikata (1999) shows that the pace of trade liberalization has quickened in South Africa since
1994 but progress has been uneven. Average protection has been lowered and although
many non-tariff trade barriers have been eliminated, several in the form of domestic dumping
and illegal customs procedures still exist. Consequently, the dispersion of effective protection
remains high and the elimination of export incentives means that the anti-export bias for
many commodities has been raised. This section focuses on the impact of tariffs and non-
tariff barriers on large manufacturing firms.

Reduction in import duties

Reduction of tariffs and duties since 1994 affected South African firms in a variety of ways.
Lower tariffs adversely affect firms through lower product prices due to foreign competition.
However, firms benefit due to a reduction in raw material prices. Of those that were
adversely affected by tariff cuts, 60 percent of size 1 firms noted experiencing a reduction in
product prices, while only 46 percent of size 3 firms said that they were similarly affected.
Although the (median) reduction in product prices subsequent to tariff cuts was in the range
of 10 to 12 percent across all firms, about 57 percent of size 1 firms compared to only 38
percent of size 3 firms reported that the cuts were significant for business. This was further
reinforced by a disproportionate loss in sales revenue for size 1 firms. While sales dropped
10 percent for size 1 and 2 firms, the corresponding drop for size 3 firms was only 3 percent.

      Of the firms affected by tariff reductions, 45 percent noted gains from a drop in
imported raw material prices. While the (median) drop in raw material prices ranged between
5 to 7 percent, a smaller proportion (19 percent) of size 1 firms compared to 23 percent of
size 3 firms classified the gains as significant.

Non-tariff Trade barriers

The effect of non-tariff barriers is positively related to firm size. They affect between 30 to
40 percent of the large manufacturing firms. Chart 5.4 shows firm ratings of the extent to
which these barriers act as constraints to firm growth. Domestic dumping by foreign firms 25
and illegal customs control procedures are the leading problems for firms. Chart 5.4 shows

25
  It is likely that firms misinterpreted the sale of cheaper foreign goods in local markets as dumping. So the
responses reported here are perhaps a result of a vague definition of dumping as defined by the firms and not an
outcome of a formal WTO-approved evaluation.
                                                                -28-


that for about 25 percent of firms, domestic dumping poses a serious problem and for another
15 percent a moderate problem. It was found to be most pervasive in textiles where 89
percent of firms are adversely affected. In other sectors, the proportion of affected firms was
closer to 40 percent. In the case of illegal customs control procedures, 20 percent find it a
serious problem while 10 percent consider it a moderate problem. Again, the share of firms
affected most by illegal imports is in textiles (86 percent). Other problems include foreign
import licensing procedures affecting South African firms’ ability to export to foreign
markets (20 percent of trading firms) and foreign anti-dumping actions preventing South
African firms from selling large volumes in foreign export markets (10 percent). Details on
how non-tariff barriers affect South African manufacturing firms’ profitability in the
domestic market and ability to trade in foreign markets are provided in Annex 6. While this
survey measures the proportion of firms affected by these barriers, the actual magnitude of
the effect remains unclear.
Chart 5.4 Firm ratings of trade barriers, 1999


                      Illegal customs control procedures


               Foreign import licensing affecting exports


          Foreign anti-dumping actions affecting exports


      Domestic dumping by foreign firms affecting profits

                                                            0   5     10   15   20   25    30   35    40    45   50
                   Serious Problem       Moderate Problem           Reasonably Efficient             Percent of firms




Special DTI programs

In recent years, the Department of Trade and Industry (DTI) has initiated a number of special
incentives for exports. This section focuses on the special programs and facilities that were
established by DTI since 1994 to assist firms in export promotion. Some of these schemes
apply to all the sectors while others are sector specific. The Export Marketing and Investment
Assistance Scheme (EMIA) aims at partially compensating exporters for costs incurred in
developing export markets. The Life Scheme provides low interest finance for exporters.
The World Player Scheme aims at improving international competitiveness in the textile,
clothing, footwear and motor vehicle sectors. The Motor Industry Development Program
(MIDP) enhances the productivity and competitiveness of the local vehicle assemblers and
component manufacturers.

      More firms use programs that affect the direct price of traded good. The two schemes
used extensively by manufacturing firms in South Africa are tax exemptions (including VAT
and customs duty) and the forward foreign exchange cover. 26 MIDP, targeted to the vehicles
sector, is used by 31 percent of the firms although as many as 52 percent are aware of it.

26
     This result is significant at the 5 percent level.
                                                                                                           -29-


Chart 5.5 shows that in general, both awareness and use of DTI programs is very low. 27
Between 50 - 78 percent of trading firms are aware of the DTI programs for tax exemptions,
forward foreign exchange cover, export marketing and investment assistance and export
credit guarantee. While 50 – 56 percent use the forward forex cover and tax exemptions
programs, the others are used by less than 30 percent of firms. The corresponding numbers
for other schemes are even lower. For instance, the Life Scheme, World Player Scheme, and
Transport subsidies are used by less than 3 percent of firms.

Chart 5.5 Awareness and use of special facilities for export, 1999

                                                                                    Use                                     Aware

                    90                                                                                                                                                                     90
                    80                                                                                                                                                                     80




                                                                                                                                                                                                Percent of firms
 Percent of firms




                    70                                                                                                                                                                     70
                    60                                                                                                                                                                     60
                    50                                                                                                                                                                     50
                    40                                                                                                                                                                     40
                    30                                                                                                                                                                     30
                    20                                                                                                                                                                     20
                    10                                                                                                                                                                     10
                     0                                                                                                                                                                     0
                          World player     Transport   Export credit      Motor industry   Export credit   Export credit       Export      Export credit   Forward forex        Tax
                           scheme          subsidies   facility - post    development      and foreign     facility - pre   marketing/      guarantee         cover         exemptions:
                                                         shipment           program         investment       shipment       investment                                     VAT / customs
                                                                                           reinsurance                      assistance
                                                                                                                              scheme




      Of the firms that use the program, managers also commented on how relevant these
programs were to their businesses. Again, as Chart 5.6 confirms, the price-oriented programs,
tax exemptions and forward forex cover, were considered essential by a third of the firms and
useful by an equal proportion. Less than 30 percent of the firms found some of the others to
be useful, even fewer deemed them as essential. The least relevant programs were World
Player Scheme, Life Scheme, and transport subsidies. The Motor Industry Program was
again an exception – rated to be essential by 60 percent and useful by 30 percent of firms.

Chart 5.6 Importance of special incentives for exports

                                                                                                            Essential                    Useful

                                            Tax exemptions: VAT / customs
                                                          Forward forex cover
                                                       Export credit guarantee
                         Export marketing/investment assistance scheme
                                          Export credit facility - pre shipment
                         Export credit and foreign investment reinsurance
                                         Export credit facility - post shipment
                                                        World player scheme
                                                          Transport subsidies
                                                                         Life scheme

                                                                                           0                         20                        40                          60                                      80
                                                                                                                                         Percent of firms aware of scheme




27
   Note, in the case of the Motor Industry Development Program, figures are based only on the total number of
firms in the relevant sector and are not shown in the charts.
                                                              -30-


      It is possible that since awareness and use are directly related, poor usage and
relevance may simply be a result of the fact that firms do not know about DTI’s special
programs and may not be an indication of the quality of the program. This possibility was
explored by asking firm mangers that use the facility to rate it with regard to its cost (Chart
5.7). The forward foreign exchange cover program and the tax exemptions programs are also
rated higher on costs and importance. 28 In general, the cost of using most programs was
considered reasonable by many firm mangers who are aware of the programs, regardless of
usage. The exceptions that were considered to have prohibitive costs were the Life Scheme
by 13 percent of firms, forward forex cover by 9 percent (possibly more by size 1 firms than
others), export credit guarantee by 7 percent and tax exemptions by 5 percent of firms. The
Motor Industry Development Program rates as the most important and most cost effective.

Chart 5.7 Firm ratings of costs of special incentives for exports

                    Tax exemptions: VAT / customs
                                Forward forex cover
                             Export credit guarantee
 Export marketing/investment assistance scheme
                 Export credit facility - pre shipment
 Export credit and foreign investment reinsurance
               Export credit facility - post shipment
                               World player scheme
                                Transport subsidies
           Prohibitive
                                        Life scheme
           Reasonable
                                                         0   10       20        30       40        50   60
                                                                     Percent of firms aware of scheme



      Since both awareness and use of programs is very low, it calls for a need to review
these programs. In this context, the experience of Morocco, which developed similar
programs successfully in collaboration with the private sector, is provided as an example in
Box 5.3.


     Box 5.3 Facilitating Industrial Clusters in Morocco

     Morocco has successfully facilitated a program where over 300 high-level participants from the private
     and public sectors worked hand in hand to develop concrete initiatives that enhance firms’
     competitiveness in the wake of new trade agreements with the European Union. These consultations
     led to some purely private initiatives, such as the creation of a quality label for sea-products and the
     setup of information services that would provide all actors in the sea-product industry with instant data
     on markets and prices all over the world. Other initiatives led to the public-private cooperation, such
     as the transformation of the Ministry of Tourism into a marketing agency jointly financed by the public
     and private sector.
     Source: New Foundations: Private Sector Development in Post-War Bosnia and Herzegovina, World
     Bank, 1997

28
     These results are all significant at the 5 percent level.
                                                     -31-


         6. LABOR REGULATIONS AND IMPLICIT COSTS
Integration with global markets requires South African firms to continually adapt to rapidly
changing global demand patterns by adjusting production technology and skills. We examine
whether labor market regulations instituted to redress workplace inequalities reduced firm
flexibility to respond to global and domestic market conditions. Since CEOs perceived labor
regulations as the second most important constraint to firm growth, we examined in detail
how firms responded to four recent labor regulations. We also looked at several unintended
consequences of labor regulations that result in raising the implicit costs of doing business
with labor. 29 Before examining the firms’ response to labor regulation, we looked at their
role in employment generation.
Role of Large Manufacturing firms in employment generation
      We measured the role of large firms in employment generation30 in GJMA as the ratio
of the average (median) number of non- managerial workers to managers per firm (Chart 6.1).
We found that size 3 firms generate more employment per ma nager in all skill categories
than size 1 or 2 firms. 31
Chart 6.1 Ratio of non-managerial workers to managers per firm



         Prof/manager

     Clerical/manager

     Service/manager

     Plant op/manager

       Crafts/manager

         Lab/manager

                          0                  2                  4                    6                 8
                                          Size1             Size2            Size3


      Among the more skills- intensive occupations, such as professionals and technicians,
the ratio of professionals or clerical workers to managers in size 3 firms is almost twice as
high as in size 1 firms. In the less skilled categories, the ratio of laborers or plant operators to
managers in size 3 firms is 1.5 times as high as in size 1 firms. 32 We also measured labor


29
   Explicit labor costs are not measured for two reasons: 1) a fairly large proportion of firms were unwilling to
reveal disaggregated wage and non-wage cost-related information; and 2) where available, the data is sparse and
not representative.
30
   Also see Annex 7 for more details.
31
   This result is significant at the 5 percent level.
32
   Survey data indicates that in GJMA, while size 3 firms create more jobs in every employment category, this
evidence in itself does not provide a case against the promotion of smaller firms. Nor does it lend support to
                                                           -32-


intensity across different firm classes. Size 1 and 2 firms employ more less-skilled workers
per unit of capital than size 3 firms. 33 While size 3 firms employ only 0.25 laborers, size 1
and 2 firms employ as many as 0.5 to 0.6.

Firm response to recent labor regulations
We examined firm responses to individual labor regulations. The labor regulations are,
Labor Relations Act (LRA) of 1995, Basic Conditions of Employment Act (BCEA) of 1997,
Employment Equity Act (EEA) of 1998 and, Skills Development Act (SDA) of 1999. Firms
were asked how they adjusted their employment levels in response to each regulation.

       Firm responses are shown in Chart 6.2. Around 65 percent of all firms reported “no
effect,” implying that each of the four recent labor regulations did not individually affect
their employment decisions. Between 20 to 27 percent of the LFS managers noted that the
LRA, BCEA and EEA led to lower employment. Firms that recorded their response as
“raised employment” represented less than 4 percent of all firms. No significant size-class
differences were observed.

Chart 6.2 Response of firms in GJMA to individual recent labor regulations, 1998


                       NA

       Not familiar with it

                No effect

      Lower employment

      Raise employment

                              0   10        20        30        40       50       60        70
                                                                                 Percent of firms

         Labor Relations Act                         Basic Conditions of Employment Act
         Employment Equity Act                       Skills Development Levy


      These firm responses according to Human Resource Managers appear to be at odds
with firm CEOs ranking of “labor regulations” as the second most important obstacle to firm
growth. This discrepancy could be attributed to the fact that it was difficult for the HR
managers to pinpoint specific labor regulations as being problematic. Therefore we
examined the cumulative firm response to all recent labor regulations combined in Chart 6.3.
Close to 40 percent of firms in GJMA noted that they adjusted to the new labor market
environment by doing one or more of the following: hiring fewer workers, substituting
machinery for labor when expanding, hiring temporary rather than permanent workers, and
relying on sub-contracting. Each of these responses has the same unintended consequence of

policies that may promote size 1 firms to the detriment of size 2 or 3 firms that, at least in the medium term, are
likely to remain South Africa’s largest employers outside the mining sector.
33
     This result is significant at the 5 percent level.
                                                          -33-


reducing permanent employment in the manufacturing sector. Firms also reported a positive
impact of recent labor regulations, with 30 percent of firms reporting improved labor
relations and 13 percent reporting improved labor productivity.
Chart 6.3 Cumulative response of firms in GJMA to all recent labor regulations, 1998


   Increased labour productivity

      Improved labour relations

           Use sub-contracting

  Hire more temporary workers

           Use more machinery

              Hire fewer workers

                                   0   10       20   30   40     50   60   70     80    90    100

        Yes           No           Don't know
                                                                                Percent of firms



Labor market environment
We examined whether other labor market factors had unintended consequences which raised
the implicit cost of doing business with labor. Factors that impede flexibility of firms to
respond to changing local and global market conditions include: number of unions that firms
do business with, level at which collective agreements regarding wages and other benefits are
struck, number of strikes experienced and workdays lost, number of disciplinary actions per
year, and hiring and retrenchment costs in time and money. We examined each of these
factors in detail and found that the results were mixed.

1. Number of unions that firms work with

      Chart 6.4 shows that the average numbers of unions that firms work with increases
with size-class. Size 3 firms deal with more unions and thus may incur higher time and
resource costs. In size 3 firms, about 44 percent do business with 1 union, 22 percent with 2
unions and 16 percent do business with 3 unions. Unlike other firm sizes, as many as 7
percent of size 3 firms work with 4 and 5 unions while 2 percent work with 7 unions. Only 1
percent has no unions. In contrast, around 82 percent of size 1 firms deal with 1 or no
unions. Hardly any size 1 firms deal with more than 3 unions. Approximately 61 percent of
size 2 firms deal with 0 or 1 union. Relative to size 1, twice the proportion (28 percent) deals
with 2 unions. A very small proportion of size 2 firms work with 3 to 5 unions.
                                                      -34-



Chart 6.4 Number of unions GJMA’s large manufacturing firms work with


        Size 1                                                                   2 unions
                                                                   Size 2          28%
                     2 unions                                                               3 unions
                       14%                                                                     5%

  1 union                   3 unions                                                           4 unions
   75%                         4%                         1 union                                 5%
                                                           56%
                                                                                                   5 unions
                            0 union                                                                   1%
                              7%                                                              0 union
                                                                                                5%




                         Size 3            4 unions
                                              7%       5 unions
                                3 unions                  7%
                                                                  6 unions
                                  16%
                                                                     1%

                                                                      7 unions
                        2 unions                                         2%
                          22%

                                                                0 union
                                                                  1%

                                                      1 union
                                                       44%




2. Collective Agreements

The level at which collective agreements are struck between employers and employees is
another indicator of business flexibility. The lowest level of agreement is reached at the level
of the plant or establishment while the highest is struck at sector or industry level. The more
aggregated the level at which employment contracts are settled, the less latitude the employer
has to reflect in the wage and benefits contract any firm-specific, worker-specific or local
environment-specific conditions. Survey data indicates that there is a positive relationship
between firm size and the level of collective agreements and that most agreements are
reached at the sector/industry level.
                                                                         -35-


Chart 6.5 Level at which collective agreements are made


                          60

                          50
       Percent of firms



                          40

                          30

                          20

                          10

                           0
                               No agreement    Establishment   Company level     Sector/Industry       Wage
                                                   level                              level        determination

                                      Size 1       Size 2       Size 3          All firms



    Chart 6.5 illustrates that about 13 percent of LFS firms are not bound by any collective
agreements. This finding is more 34 applicable to the size 1 and 2 firms. Among size 3 firms
that employ over 75 percent of the manufacturing workforce, only 3 percent of the firms are
free of collective agreements. About 12 percent of firms strike collective agreements at the
establishment or plant level, and 26 percent at the level of the company. There is no
significant difference across size-classes. At the highest level, the extent of collective
agreements rises to include about 46 percent of the firms. Overall, about 4 percent of firms
are subject to Wage Determination Board decisions.

    By constraining a firm’s ability to determine wages on the basis of local market
conditio ns, a system of collective bargaining that occurs above the plant level can raise the
relative price of labor and adversely impact employment. Approximately 21 percent of the
managers surveyed reported that collective Bargaining Council System led to a decline in
employment, and 5 percent reported a rise in employment as a result. However, 59 percent
reported no impact. 35

3. Number of strikes, lockouts, stay-aways and workdays lost in 1998

Table 6.1 shows the number of strikes experienced by LFS firms in 1998. Overall, 58
percent of LFS firms did not have any strikes in 1998, 30 percent experienced at least one
strike and about 8 percent of the firms experienced 2 strikes. Another 3 percent of firms
experienced more than 3 strikes. Strikes were less of a problem among size 1 than size 2 and
3 firms. Size 1 firms seemed to enjoy a relatively “hassle free” work environment with
respect to strikes and lockouts.

      Only 24 percent of LFS firms reported the number of workdays lost during strikes in
1998. This information is a useful indicator of forgone production. As shown in Table 6.2,
about 47 percent of these reporting firms lost between 1 to 10 workdays. The proportion of
firms affected is positively related to size-class; 53 percent of size 3 compared to 31 percent

34
     All comparisons hold at the 5 percent level of significance.
35
     For details by size class, see Annex 8.
                                                -36-


of size 1 firms lost between 1 to 10 days of production in 1998. Another one third lost
between 11 to 25 workdays, while about a fifth lost 26 to 200 workdays.
Table 6.1 Number of strikes, 1998 (Percent of firms)

    Number of        Size 1           Size 2           Size 3     All firms
      strikes
   0                  75.0            52.0               46.0          58.0
   1                  17.0            40.0               35.0          30.0
   2                   8.0             3.0               12.0           8.0
   3                   0.0             1.0                6.0           2.0
   4                   0.0             0.0                1.0           0.2
   5                   0.0             3.3                0.0           0.9
   25                  0.0             0.0                0.3           0.1

Table 6.2 Number of workdays lost in GJMA due to strikes, 1998 (Percent of firms)

   Number of days             50-99            100-199          200+          Total
   1 –10                       31.4              48.7           52.7          46.8
   11 – 25                     51.9              24.4           28.8          32.5
   26 – 200                    16.6              26.9           18.4          20.6


4. Disciplinary inquiries per year

Overall, LFS data shows that 22 percent of firms conducted disciplinary actions against more
than 1 in every 10 employees per year. No size-class differences were observed. The
incidence of disciplinary inquiries held for less than 1 in every 10 workers in a firm was
related to size-class. It was 62 percent on average. Numbers rose from 56 percent for size 1
or size 2 firms to over 76 percent for size 3 firms. Only 6 percent of size 3 firms but as many
as 23 percent of size 1 firms had no inquiries in 1998.


5. Time and costs of hiring and retrenchment

Adding to the cost of doing business with labor are the time and financial costs of hiring and
firing an entry-level worker. Table 6.3 shows that the average time taken to retrench an
entry- level worker is approximately 2.7 months. Approximately 29 percent of firms took
between 0 to 1 month to retrench a worker in 1998. Another 56 percent of firms took 2 to 3
months. However, it took as long as 4 to 7 months to retrench an entry level worker for 13
percent of firms. This could serve as a negative signal for potential firms and existing
employers contemplating business expansion.
                                                             -37-


Table 6.3 Time taken to retrench an entry -level worker (Percent of firms)

                                        Size 1               Size 2             Size 3       All firms
        0 months                          0                    3                  3              2
        1month                            39                   36                 5              27
        2 months                          25                   30                 57             37
        3 months                          17                   20                 21             19
        4- 7 months                       14                   11                 14             13
        16 months                         5                    0                  0              2

      Chart 6.6 graphs the median Rand costs of hiring or firing an entry- level worker in
semi- and relatively low-skills occupational categories such as crafts and related trades, plant
and machine operators and laborers and related skills.

Chart 6.6 Hiring and firing costs of an entry-level worker, 1998

          Firing costs-Laborers

  Firing costs-Plant &Mach. Op.

          Firing costs-Craft etc.

          Hiring costs-Laborers

  Hiring costs-Plant & mach.op.

          Hiring costs-Craft etc.

                                    0   2000     4000   6000        8000   10000 12000 14000 16000

        Size 1                 Size 2               Size 3                                   Rands




      There is a difference between size 1 and 3 firms with respect to the hiring costs of
craftsmen and plant operators. At R 9000 per entry-level worker, size 3 firms incurred more
than twice the cost in 1998 as size 1 firms. Size-class variations for laborers, the least skilled
occupational category are fewer and range from R 2160 – 2900. Retrenchment costs for all
three skill categories show little variation for size 3 firms. But there is a strong size-class
difference across occupations/skills. For plant and machine operators and laborers, the firing
costs per entry level worker more than double between size 1 and size 3 firms from R 3250 in
size 1 firms to R 11,600 in size 3 firms.

Temporary work and subcontracting
Firm responses to recent labor regulations resulted in the loss of a few permanent jobs but
they also created some non-permanent jobs through temporary work and sub-contracting. In
keeping with global trends, South African firms seem to be relying on temporary workers
and outsourcing. The proportion of firms that currently create jobs by employing temporary
workers and sub-contractors is shown on Chart 6.7.
                                                                               -38-


Chart 6.7 Use of temporary workers and subcontractors, 1998

      Percent of firms   100
                          80
                          60
                          40
                          20
                           0
                                          Subcontract                        Use temporary labor

                                             Size 1            Size 2         Size 3




      There is a positive relationship between firm size and the use of temporary labor and
subcontractors. About 83 percent of size 3 compared to 63 percent of size 1 firms employ
temporary workers. Chart 6.8 shows the primary reasons for using temporary workers. About
77 percent of firms rated “provides flexibility – can expand workforce without hiring
permanent workers” as the most important reason to employ temporary labor. Another 20
percent rated the same reason as being moderately important. About 17 percent of firms rated
“reduce permanent workforce” as the most important reason, another 40 percent rated it
moderately important. Only 5 percent of firms rated “lowers the fixed non-wage cost” as the
most important reason, and another 20 percent rated it as a moderately important reason for
hiring temporary workers.

Chart 6.8 Reasons for using temporary workers, 1998


                                             Other

     Reduce permanent workforce

                               Provides flexibility

      Lowers fixed non-wage costs

                                                      0   10       20   30      40     50    60    70       80      90    100

                                                                                                                 Percent of firms


                                Most important            Moderately important              Not important




      Sub-contracting of various economic activities from one firm to another is popular
among large manufacturing firms. Chart 6.7 ind icates that the percent of firms sub-
contracting various activities varies from 62 percent in size 1 firms to almost 90 percent in
size 3 firms. 36


36
     The survey did not record the percentage of an activity that is subcontracted by a firm.
                                                               -39-


Chart 6.9 Tasks outsourced by firms, 1998


                Other
             Training
          Production
    General Services
           Marketing
           Transport
   Plant maintenance
       Administrative

                        0         10         20           30           40            50        60        70
                                                                                     Percent of firms

      Chart 6.9 shows the main economic activities subcontracted by m       anufacturing firms.
   The three most common activities outsourced by 50 percent or more firms are general
   services, transport and training. Size 1 firms tend to out-source more marketing activities
   but less of general services and training compared to size 2 or 3 firms. About 45 percent
   of firms outsource production itself. Chart 6.10 shows that the leading reason for
   outsourcing production activity was “flexibility to respond to surges in the workload”
   listed by 80 percent of firms that rated it “extremely” to “moderately” important. Other
   reasons were “subcontracting is cheaper,” reported by about 53 percent of firms,
   “minimizes labor discipline” listed by 39 percent of firms. “Respond to temporary need
   for a specific skill,” a typical reason for subcontracting was listed by about 73 percent of
   firms.

Chart 6.10 Reasons for subcontracting production activity, 1998

                                             Other

           Subcontracters produce better quality

     When equipment malfunctions/breaks down

            Minimizes labour discipline problems

    Respond to temporary need for specific skills

                        Subcontracting is cheaper

      Flexiblility to respond to surges in workload

             Extremely important                      0   10   20     30   40   50   60   70   80   90 100

             Moderately important                                                     Percent of firms
             Not important
                                                                             -40-


                                               7. AVAILABILITY OF SKILLS
Shortage of skills was identified as a major constraint to business growth by 25 to 30 percent
and a moderate constraint by 32 to 55 percent of CEOs of manufacturing firms. Supporting
evidence gathered by interviewing Human Resource managers shows that shortage of skills
may be more acute than perceived by the CEOs. Furthermore, if firms found it difficult to
find skilled people in a year when the economy was in near recession, it is likely to be a far
more binding constraint when growth occurs at a more rapid pace. Since shortage of skills is
likely to a binding constraint for existing investors in the short to medium term and may be a
deterrent to potential investors, we examine the issue in detail. Given South Africa’s
apartheid past, we begin by looking at the racial profile of skills in GJMA.
Racial distribution of skills employed in GJMA
In 1998-99, the distribution of skills in GJMA was closely tied to race. Chart 7.1 shows that
whites occupy 80 percent of managerial and 70 percent in professional and technical
positions. In occupations requiring medium skills such as clerical and other services
(accounting, le gal, marketing, financial etc.), there was again a disproportionately large
representation of whites at about 65 percent and a weaker representation of Africans at 24
percent. In semi-skilled occupations such as crafts, the Africans exceed whites at 50 percent.
Further down the skill ladder, among plant operators, the share of Africans is even higher at
81percent; it rises to over 91 percent for laborers.

Chart 7.1 Distribution of skills by race in GJMA, 1998


                          100
                           90
                           80
       Percent of firms




                           70
                           60
                           50
                           40
                           30
                           20
                           10
                            0
                                Managerial    Prof &       Clerical      Service        Craft        Plant op.   Laborers
                                             Technical

                                                         Africans     Coloreds     Asians   Whites




Difficulty in finding skills
According to the Human Resource Managers of firms surveyed, the shortage of skills
indicated by a statistically 37 ordered listing from “most to least difficult to find” skills was as
follows: 1) managerial, professional and other technical skills; 2) service-related skills such

37
     All results significant at 5 percent level unless otherwise stated.
                                                                           -41-


as accounting, marketing, financial, legal etc. and craft and related trades; 3) plant and
machine operators; 4) clerical skills; and 5) laborers (see Chart 7.2). About 82 percent of
firms reported extreme to moderate difficulty in finding these skills. Almost 70 percent of
firms indicated a skill shortage of either an extreme or moderate type with respect to service
and craft-related skills. In the case of semi-skilled workers such as plant or machine
operators, the difficulty of finding skilled workers drops considerably. Over 50 percent of the
firms did not have any problem finding these skills. Only 5 to 8 percent noted an extreme
difficulty. Firms did not encounter much difficulty in finding clerical workers and laborers.

Chart 7.2 Difficulty in finding skills, 1999

                        100

                        80
     Percent of firms




                        60

                        40

                        20

                         0
                              Managerial   Professional      Clerical        Service      Craft      Plant operator   Unskilled

                                            Extremely hard              Moderately hard       Not hard



      Across the three size-classes, differences in the difficulty of finding skills were distinct
only in the case of managerial, professional & technical and service-related skills. In
particular, relative to size 1, size 3 firms fo und it more difficult 38 to find highly skilled
workers.

Vacancy rates and levels

Since firms reported difficulty in finding skills, we examined vacancy rates by skill. At least
75 percent of firms report no vacancies and this percentage increases as the skill level
declines. A significantly39 greater proportion of size 3 firms recorded non- zero vacancy rates
for highly skilled workers than size 1 or 2 firms. These vacancy rates prevailed in as many as
30 percent of firms after a period of continued lay-offs. Vacancy rates for medium- level
skills, such as clerical or service-related skills, are more pervasive among size 3 firms. 40
Around 20 percent of size 3 firms compared to less than 10 percent of size 2 firms reported
vacancies. Among size 1 or 2 firms, vacancy rates for semi-skilled workers are roughly
similar to those for medium skilled workers. In descending order, HR managers cited the
following as the main reasons for existing vacancies: 1) can’t find the right person; 2)
expansion; 3) dismissal; 4) resignation; and 5) retirement

38
   Size 3 firms were distinct from size 2 and size 1 firms at the 5 percent level in the case of managerial and
professional & technical skills; and at the 10 percent level in the case of service-related skills.
39
   Among the firms reporting non-zero vacancy rates, the proportion of size 3 firms is statistically greater at the
5 percent level.
40
   Among the firms reporting non-zero vacancy rates, the proportion of size 3 firms is statistically greater at the
5 percent level.
                                                          -42-



Recruitment and hiring preferences
The preceding discussion revealed a scarcity of highly skilled workers. In recent years, there
is growing concern about providing job opportunities for workers with semi and low skills,
hereafter referred to as production workers. Consequently, a large section of the survey
focused on firms’ recruitment preferences for production workers and the human capital
accumulation practices that are underway to redress the shortage of technically skilled
workers in the labor market.

    For production workers that constitute the largest occupational skills category in a
manufacturing firm, we asked the Human Resource Managers to rank their recruitment
preferences for three age groups: 16-24, 25-45 and 46 and above. Approximately 72 percent
of all firm managers in GJMA noted that they preferred to hire production workers in the age
group of 25-45 (see Annex A9). Only 9 percent of firm managers expressed a recruitment
preference for production workers in other age groups. This preference applied across all
size-classes. 41

    In hiring production workers, 54 percent of firms seemed indifferent to gender, 34
percent preferred males while 12 percent preferred females. There were no significant 42
differences in this pattern across size-class. One of the main reasons for preferring male to
female workers was the nature of the work in certain sectors such as metal products or
vehicles and auto parts.

   We asked the HR managers to rank their preferences for the qualifications of production
workers in terms of the following:
   a) Relevant work experience only
   b) Education only – primary, secondary, vocational or tertiary
   c) Relevant work experience and education:

                i.    Relevant work experience and primary education
               ii.    Relevant work experience and secondary education
              iii.    Relevant work experience and tertiary education
              iv.     Relevant work experience and vocational education

      Survey evidence shows that 75 percent of the HR managers preferred job applicants
with relevant work experience. 45 percent of the HR managers preferred applicants with
secondary education and 26 percent preferred vocational education. Applicants with primary
and tertiary education were preferred by 10 –12 percent and not preferred by 30 – 33 percent
of the HR managers (see Annex 9). Among applicants with a combination of work
experience and education, the qualifications most preferred by over 66 percent of HR
managers were “work experience plus secondary education.” The second most-preferred
category for 42 percent of managers was “work experience plus vocational education.” The

41
     This result is significant at the 5 percent level.
42
     This was true at the 5 percent level.
                                                                                 -43-


least preferred combination for 20 percent of managers was “work experience plus tertiary
education.”

Chart 7.3 Recruitment preference for production workers in GJMA (Number of firms)


                        1800

                        1600
 Number of GJMA firms




                        1400

                        1200
                        1000
                         800

                         600
                         400
                         200

                           0
                               Age (25-45) Either male   Relevant    Secondary          Prim. educ   Sec. educ   Voc. educ   Tertiary edu
                                            or female    work exp,   educ only          plus work    plus work   plus work    plus work
                                                           only                            exp.         exp.        exp.         exp.


Note: Chart 7.3 is constructed using regional weights; therefore, the vertical axis shows the number of firms in GJMA (out
of a total of 2674 firms).

      Chart 7.3 shows that for the age group of 25 to 45, a person of either sex is most likely
to be considered a preferred candidate if he/she has a combination of secondary education
and work experience. This poses a potential problem for the large pool of presently
unemployed who have some education but no work experience. According to HR managers’
recruitment preferences, such workers are only “weakly employable.” Specifically, firm
managers have a clear preference for high and medium skilled workers, whereas the present
supply of workers seems to be characterized by low skills. Likely candidates for the category
of weakly employable are: unemployed youth who entered the labor force in the 1990s and
never held a job before, and workers with skills that are not very relevant for manufacturing
firms. An example of the latter is retrenched mineworkers.
Worker training
       While government sought to legislate the Skills Development Act of 1998 that allows
firms to claim a reimbursement against training costs, the private sector has introduced its
own initiatives to accelerate worker training. This section examines the nature of worker
training undertaken by large manufacturing firms in GJMA.

      LFS evidence shows that between 36 to 47 percent of firms provide in- house training
for managerial/professional, semi-skilled and low skilled workers with large sectoral
variations. Outside training for managerial, professional and semi-skilled workers is provided
by 37 to 42 percent of firms and for low-skilled workers by 9 percent of firms. While on
average, about 33 percent of manufacturing workers in firms with over 50 employees receive
some type of training, there are strong sectoral variations. In the set of firms that undertook
any type of training in 1998, the largest proportion of workers trained varied from 65 percent
                                                     -44-


in chemicals, 41 percent in vehicles and parts to 11 percent in textiles (See Annex 9 ). Firms
were reluctant to reveal how much they spend on training. 43

Type and magnitude of training

The LFS indicates that firms undertake in- house and outside programs designed for workers
in a particular skill category such as highly-skilled (managerial and professional), semi- or
low skilled, or joint training programs serving several skill categories. Of the firms that
provide training, the median number of workers trained per firm provides a good measure of
the coverage of training currently provided. Table 7.1 shows that in 1998, the median number
of workers trained per firm in both in-house and outside programs varied from 17 in highly-
skilled, 24 in semi-skilled, to 40 in low-skilled categories.

In-house training : Table 7.1 sho ws that less than half of the LFS firms provide in- house or
outside training. In-house training is provided by about 36 percent of the firms for high- and
low-skilled workers and by 48 percent of firms for semi-skilled workers. Common or joint
in- house training that covers all or most skill categories is offered by 17 percent of the firms.
The median number of workers trained varied from 9 for the managerial/professional
category to 16 for semi-professionals and 20 for unskilled workers with significant sectoral
variations. As a share of total employment, this number appears low for the large firm
manufacturing tier as a whole. 44 In this tier, the smallest size-class employs between 50 to 99
employees and the largest size-class employs over 200 employees. In fact, the largest firm in
the LFS had over 10,000 employees.

Outside training programs : Table 7.1 indicates that less than half the LFS firms use outside
training sources. About 42 percent of firms engage in outside training for managers while
another 37 percent train semi-skilled workers. For low-skilled workers, the proportion of
firms that use outside sources is 9 percent. The numbers for joint courses is 8 percent. The
median number of workers trained per firm varies from 8 for high and semi-skilled workers
to 20 for low-skilled workers. Sectoral details are provided in Annex 9. 45




43
   Among the various types of training programs, the most difficult one to measure is on-the-job training.
Because it is firm specific in nature, its coverage naturally extends to all firms and all workers but is not
captured in this survey. The remainder of this chapter focuses on in-house and outside training programs
provided by the firm in 1998.
44
   Annex 9 provides a comparison with firms in Malaysia where the proportion of firms that train rises with firm
size.
45
   HR managers of GJMA manufacturing firms were asked to rank the sources of outside training they used in
1998. The most popular sources are private training schools used by 70 percent of firms and industrial training
boards and vocational training schools/technikons used by 55 percent of firms. Universities and other business
partners were used by about 40 percent of firms. Government institutes came in last and were used by only 35
percent of firms.
                                                    -45-


Table 7.1 Worker training by GJMA firms, 1998

                                 Managerial/    Semi-skilled   Low skilled   Common/all
                                 professional                                skills

In-house training
% of firms that offer in-house       36.0           47.3           37.9         17.4
training
Average number of training            5.2            5.3           3.1           5.4
programs offered in 1998
                                      (2)            (2)            (1)          (2)
         46
Number of workers trained              9             16             20           51
Outside training
% of firms that offer out-side       41.7           36.5           8.8           8.1
training
Average number of training            5.3            6.9           3.8           8.9
programs offered in 1998
                                      (3)            (2)            (2)          (2)
Number of workers trained              8              8             20           28


Do trained workers leave after training?
As with any other public good, it may be argued that firms are likely to under- rather than
over-train workers because of the likelihood that newly trained employees will be enticed
away by other firms for higher wages. The LFS indicates that 36 to 47 percent of firms
provide in- house or outside training programs for about 33 percent of their workers.
However, despite the scarcity of skilled workers, LFS data provides no evidence to support
the hypothesis that firms train a relatively small number of workers because workers leave
after training. Less than 5 percent of the trained workers leave after training. It is possible
that firms may not have been losing trained workers to other firms because the economy was
sluggish at the time of the survey.




46
     Numbers given are medians.
                                             -46-


        8. LICENSING REGULATION AND CORRUPTION

Complicated regulatory requirements are often seen as imposing pecuniary and non-
pecuniary costs on firms. In South Africa, regulatory requirements do not appear to be
problematic. Licensing and permit requirements for either starting a new business or
continuing operations do not pose problems for a large proportion of LFS firms. The legal
managers of over 75 percent of firms surveyed reported that the number of licenses and
permits required to start-up operations and to continue them were reasonable. In addition to
regulations, CEOs list corruption in government as obstacle to firm growth. While corruption
can assume a number of forms, the survey focused on the issue by exploring whether the
regulations and procedures that were required for obtaining government’s licenses and
permits were corrupt. CEO perceptions do no t hold up under closer scrutiny.
Start-up licenses and permits
On average, size 3 firms required 5 licenses and permits in 1998 to start a business whereas
size 1 and 2 firms required only 2. Approximately 41 percent of all LFS legal managers
responded to the question regarding difficulties in obtaining licenses to start operations. Only
7 percent of these managers reported that the number of various permits, licenses and other
regulatory requirements needed to start a new business was excessive, thus confirming that
the number of permits required to start an operation was not a significant problem for most
firms.

Table 8.1 Firms that found regulations costly for starting a business (Percent of firms)

                              Size 1           Size 2           Size 3           All firms
 Time costs – very costly     15               13               11               13
 Money costs – very costly    5                5                4                5
 Any other costs/bribes –     0                0                1                0.4
 very high

       Table 8.1 records the proportion of legal managers who found existing regulations for
starting a business very costly. Clearly, time costs are more of a problem than the financial
costs or fees: while 13 percent of the firms complained about the time taken to obtain the
licenses, only 5 percent griped about their financial burden. Additionally, size 1 firms were
marginally more affected than size 2 or 3 firms. Only 1 percent of size 3 firms complained of
other types of costs (such as bribes) that the business regulatory system imposed on them.
Licensing and permits to continue operations
Approximately 52 percent of legal managers answered the question on difficulties in
obtaining licenses to continue operations. On average, the number of licenses needed to
continue operations by size 1 firms is 1, by size 2 firms is 2 and by 3 firms is 3. Similar to
the case of start-up licenses only 6 percent of the managers noted that the costs of obtaining
licenses were excessive. The remainder found the costs of continuing operations reasonable
or not an issue.
                                                  -47-


      In many countries, in addition to bureaucratic hassles and excessive license
requirements, firms are expected to pay bribes to obtain services. Only 5 percent of the LFS
managers reported that they were expected to pay bribes to obtain or expedite licenses to
continue operations. Table 8.2 compares the time costs versus the money costs and bribes
that LFS firms inc ur. As in the case of start up licenses, percent of firms complaining about
time costs is higher than those that find financial costs and bribes a problem. About 13
percent of the legal managers rate time costs as “very costly” relative to only 5 percent who
complained about money costs and 1 percent who noted bribes as problematic.

Table 8.2 Firms that found regulations very costly for continuing operations (Percent of
firms)

                                     Size 1        Size 2      Size 3            All firms
 Time costs – very costly            12            15          12                13
 Money costs – very costly           6             5           3                 5
 Any other costs/bribes –            1             1           1                 1
 very high

       Although a small proportion of managers noted business regulations as problematic,
the LFS examined how the firms were affected. Only 4 percent of LFS firm managers
reported a change in firm response due to the licensing and permit requirements. Of those
that noted a change in response, most changed the pattern of employment from full- time to
part-time.
Cumbersome procedures and corruption
Besides examining whether government’s licensing procedures imposed unduly high costs on
the firms’ ability to operate, the LFS also asked administrative and legal managers to identify
government departments whose procedures and regulations are found to be particularly
cumbersome.

Chart 8.1 Firms that find government procedures cumbersome


                                 Other

      Local authorities of Johannesburg

                Department of Finance

                  Department of Health

                                   DTI

                  Department of Labor

                                          0   5          10   15        20         25        30

                                                                         Percent of firms
                                             -48-


     Chart 8.1 reveals the departmental ratings assigned by the legal managers on the basis of
 the complexity of procedures. On average, in descending order, the ratings of government
 departments is as follows: at the national level, the Department of Labor was rated as the
 department with the most cumbersome procedures by about 28 percent of firms, and the
 Department of Trade and Industry was rated as the department with the second most
 cumbersome procedures by 23 percent of firms. At the local level, GJMC was rated by 24
 percent of the firms as having some of the most cumbersome procedures. Given that almost
 a quarter of the firms assigned these rankings to government departments suggests that there
 is room for improved regulatory efficiency and rationalization of procedures at both the
 national and local levels.

    Focusing on corruption, the LFS asked firms to report on whether they were required to
 pay bribes. As noted earlier, no more than 1 percent of firms reported that they were asked to
 pay bribes.
 Government contracts
 The frequency with which firms do business with government is evaluated in the survey in
 terms of the process through which government contracts are awarded. About 30 percent of
 firms applied for government contracts in 1998. It took 3 months on average for the tenders
 to be approved.

 Table 8.3 Median number of government tenders submitted and approved, 1997 and
 1998

                              1997                                   1998
                 Size 1      Size 2       Size 3        Size 1       Size 2       Size 3
Submitted          1           5            6             1            5            6
Awarded            0           1            5             0            2            5
Not awarded        1           4            1             1            3            1

 Table 8.3 shows the number of tenders submitted and approved in 1997 and 1998. On
 average, more tenders were awarded to size 3 firms than size 1 or 2 firms.
                                                          -49-



            ANNEX 1. LOCAL ECONOMIC DEVELOPMENT
                        METHODOLOGY
The LED methodology47 focuses on micro-evidence from the real sectors to inform policy-
makers at the local and national levels. The LED methodology aims at reducing poverty
through (1) sustained employment and income creation (referred to as income poverty
reduction) opportunities, and (2) improved social service delivery (referred to as non- income
poverty reduction) (see Diagram A1.1). Income poverty reduction through sustained
employment creation can be promoted by informed policies based on a rigorous
understanding of the current demand and supply of labor at the level of firms and workers.
The LED methodology described in Diagram A1.1 and A1.2 necessitated a set of 6 surveys
conducted in the Greater Johannesburg area.

Identifying the missing gaps

Informed policy making required a comprehensive, updated economic database in GJMA.
The following questions needed to be answered by such a database (See diagram A1.2 for
details).

1. What is the nature of economic activity at the sectoral level in GJMA? The emphasis was
   on private sector activity in manufacturing, trade, services, construction, etc.
2. Where is economic activity located in GJMA? This question relates to the spatial
   dimensions of economic activity. Are the sources of growth and employment in the
   Central Business District? In the suburbs? In the industrial districts? In townships or in
   informal settlements? Is the pattern of growth and employment concentrated in some or
   all of these? How are these areas, especially the previously disadvantaged ones, linked to
   the larger markets? Is there a disconnect between labor markets and production centers?
3. How is economic activity organized across the various tiers of the industrial spectrum?
   This relates to the vertical nature of economic activity as it occurs in the large formal tier,
   SMMEs and informal micro-enterprises? Are these tiers economically linked as in
   subcontracting from the large formal firms to the formal SMMEs or the informal
   SMMEs? Are SMMEs outsourcing to the micro-enterprises in the informal sector? Is
   informal output a substitute or a complement to formal output (good or service) at the
   sectoral level?
Tools to fill the missing information gaps

National level firm data was available but it was neither disaggregated at the level of the
large metropolitan areas, nor was it current to facilitate detailed sectoral analysis. To fill the
missing gaps, the GJMC study used the following instruments:

1. Formal sector survey of large (with 50 or more employees) manufacturing firms (325
   firms) across 8 manufacturing sectors.

47
   The methodology for non-income poverty reduction or improved social service delivery also evolved from the
Johannesburg study pilot and is ongoing. It is described in a separate report and is not a part of the 6 report LED study.
                                                -50-


2. Formal sector survey of large (with 50 or more employees) service sector firms (175
   firms) across tourism and IT sectors.
3. Survey of SMMEs (with less than 50 employees) in the formal sector (880 firms) across
   8 manufacturing and service sectors.
4. Survey of informal sector firms (500 firms) in GJMA.
5. Survey of 1040 households to understand the role of human capital, social capital and
   institutions in Soweto. The motivation for this survey is to understand, at the leve l of the
   household and local communities in townships and inner city, the weaknesses of existing
   institutions that promote or constrain human and social capital accumulation. This
   information will be useful in identifying the scope for government and private sector
   initiatives which can help communities to mobilize savings/resources at the household
   level for local investment, fight the war against aids and crime, community initiatives
   which can improve the quality of schooling/skills formation in township /rural schools,
   local initiatives to better disseminate market information to promote small businesses.
6. Survey of all training and credit providers financed by Khula and Ntsika in GJMA.
7. Worker survey of 400 recently hired production workers.

                                              2.1
                                     Diagram A1.1
                                  Goals and Objectives

                              Goal of LED : Poverty Reduction




                         Income poverty                     Non-income
                         reduction                        poverty reduction




                   Through employment                        Better quality of life - service
                      creation - LED                                    delivery

                                                           Understanding demand and supply
               Understanding demand and
                                                              of services in households
                supply of labor in firms



                                   Monitoring Jobs & Firms - RSC
                               Monitoring service delivery in households
                                              -51-


                                        Diagram A1.2

                            Economic information system in GJMC




 What is the nature of                Where is it located? What       How is it distributed across
 economic activity in                      are its spatial             the industrial spectrum?
        GJMC?                             characteristics?                 Vertical nature ?
 Sectoral distribution?




     Manufacturing?                          Are the centers of            Large firms?
       Mining?                            production in the CBD?      Small, medium & micro
       Services?                                Inner city?                   firms?
        Trade?                                Townships?                 Informal firms?




                                            Are these centers of           Are the tiers of the
                                         production self-contained?        industrial spectrum
 Constraints? Enabling                     Stagnant? Emerging?                linked? Is their
 factors? Employment                        Inter-linked through           relationship one of
potential of each sector?                           trade?                   substitution? Or
                                                                            complementarity?
                                                 -52-


              ANNEX 2. SAMPLE DESIGN AND WEIGHTS

Sampling technique

Approximately 85 percent of large firms are located in metropolitan areas (Chart A2.1). In
the next section, the technical details of the sample are presented and the weighting issues
arising from sample design and their implications for data analysis are addressed. Although
the weighting process is not error free, it manages to remove most of the design effects.

Chart A 2.1 National distribution of Large Manufacturing Firms across provinces, 1999

        Free State Mpumalanga                        N. Province       Northern Cape
           4%         4%                                 3%                 1%
                                       North West
                                          3%
     Kwazulu-Natal                                                              Western Cape
         18%                                                                        17%




                                                                                Eastern Cape
                                                                                     7%
                                             Gauteng
                                              43%

Sample design

Although the 1999 large firm survey in the Greater Johannesburg metropolitan area (GJMA)
is extensive in its coverage, it is not a census. The eight manufacturing sub-sectors in the
survey were assigned national sample sizes of 100. The survey was further stratified by
employment class (small: 50-99 employees, medium: 100-199, and large: 200 and above).
Stratification by employment class within the different sectors was proportional to size.
Within these multi-strata, simple random sampling was performed. This sample design
introduces statistical effects 48 that must be removed using appropriate weights.

Sample frame

We began with the complete frame of large manufacturing firms in South Africa and in
GJMA proper. The frame of large manufacturing firms in South Africa is listed in Table
A2.1.



48
  Though the sample design is not as complex as in most LSMS surveys where two-stage cluster sampling is
performed, it is however more complex than a simple random sampling design.
                                           -53-


Table A2.1 Manufacturing firms in South Africa by sector and size

                                         200 +        100-199         50-99        Total
 Chemical Products                        293           297            432         1022
 Electrical, electronic machinery         244           220            374          838
 Food processing and beverages            361           273            393         1027
 Iron and steel                            52            34             43          129
 Metal products                           166           212            501          879
 Paper and furniture                      291           271            405          967
 Textiles                                 389           281            317          987
 Vehicle, automotive components           115            83            127          325
 Total                                   1911          1671           2592         6174

The complete frame of large manufacturing firms in GJMA is listed in Table A2.2.
Table A2.2 Sample frame: Johannesburg and its surrounds by sector and size

                                         200 +        100-199         50-99        Total
 Chemical Products                        151           149            195          495
 Electrical, electronic machinery         128           127            235          490
 Food processing and beverages             89            44             91          224
 Iron and steel                            23            15             26           64
 Metal products                           101           113            267          481
 Paper and furniture                       75            77            159          311
 Textiles                                  50            43             71          164
 Vehicle, automotive components            39            27             51          117
 Total                                    656           595           1095         2346

Table A2.3 shows the new sample sizes in Johannesburg and its surrounds after reducing the
population to Gauteng, KwaZulu Natal, Western Cape, and Eastern Cape.
Table A2.3 New sample sizes in Johannesburg and its surrounds by sector and size

                                         200 +        100-199         50-99        Total
  Chemical Products                        17           17              22           56
  Electrical, electronic machinery         17           17              31           65
  Food processing and beverages            12            6              13           31
  Iron and steel                           22           14              25           61
  Metal products                           13           15              35           63
  Paper and furniture                       9           10              20           39
  Textiles                                  6            5               8           19
  Vehicle, automotive components           13            5              17           35
  Total                                   109           89             171          369

Questionnaires were sent to firms shown in Table A2.3. Approximately 90 percent of the
overall target was attained and there were 15 firms with less than 50 employees that
responded. However, the desired quotas were closely matched as shown by Table A2.4.
                                            -54-


Table A2.4 Actual sample obtained

                                           200 +           100-199             50-99           Total
 Chemical Products                          11               16                  21              48
 Electrical, electronic machinery           17               10                  29              56
 Food processing and beverages              11                6                   9              26
 Iron and steel                             18               13                  25              56
 Metal products                              9               18                  30              57
 Paper and furniture                        10               12                  12              34
 Textiles                                    3                6                   5              14
 Vehicle, automotive components             13                7                  14              34
 Total                                      92               88                 145             325


Determination of weights

The weights at the regional and national level were obtained as the inverse of the actual
sampling fraction. Tables A2.5 and A2.6 show the national and regional weights by sector
and size-class. Note that most of the regional weights are approximately 10 or less. The
only exception is the iron and steel sector where a more complete enumeration was done due
to the smaller number of firms in this sector. The other exception is the textiles sector.
Table A2.5 Regional weights

                                                   200 +             100-199           50-99
 Chemical Products                                 13.73               9.31             9.29
 Electrical, electronic machinery                   7.53              12.70             8.10
 Food processing and beverages                      8.09               7.33            10.11
 Iron and steel                                     1.28               1.15             1.04
 Metal products                                    11.22               6.28             8.90
 Paper and furniture                                7.50               6.42            13.25
 Textiles                                          16.67               7.17            14.20
 Vehicle, automotive components                     3.00               3.86             3.64

National weights highlight the sectors that do no t feature prominently in Greater
Johannesburg. Worthy of notice is the textiles sector. The food sector is also less prominent
in Gauteng/Pretoria.
Table A2.6 National weights

                                               200 +             100-199               50-99
 Chemical products                             26.60              18.60                20.60
 Electrical, electronic machinery              14.40              22.00                12.90
 Food processing and beverages                 32.80              45.50                43.70
 Iron and steel                                 2.90               2.60                 1.70
 Metal products                                18.40              11.80                16.70
 Paper and furniture                           29.10              22.60                33.80
 Textiles                                      129.70             46.80                63.40
 Vehicle, automotive components                 8.80              11.90                 9.10
                                                             -55-


                                       ANNEX 3. DEMOGRAPHICS

Excess capacity and labor shifts

Invariant of firm size or sector, we found that approximately 80 percent of firms reported
operating at less than full capacity in both 1997 and 1998. Table A3.1 shows the additional
output that firms could produce with existing resources if demand were to increase. A
sectoral analysis showed that firms in the chemicals sector could almost double their output if
sufficient demand existed.

Table A3.1 Additional output firms could produce with existing capacity if demand
were to increase (mean percent of output by firm size)
                          1997                        1998
 Size 1                   46.5                         42.8
 Size 2                   42.6                         46.5
 Size 3                   29.5                         33.9

      We found that almost 60 percent of LFS firms used a single labor shift in both 1997
and 1998, signaling relatively high unutilized capacity levels. Chart A3.1 illustrates that less
than one third of the firms used a double shift in both years and less than 10 percent of firms
used a triple shift.

Chart A3.1 Types of labor shifts


                              70
                              60
           Percent of firms




                              50
                              40
                              30
                              20
                              10
                               0
                                   Single          Double           Triple

                                            1997            1998




Distribution of firms by capital stock and turnover

      On the assumption that firms within a sector and size-class have approximately the
same production technology, it is possible to construct a measure of the capital- labor ratio.
Table A3.2 lists the capital- labor ratio based on the mean value of a firm’s fixed capital stock
per employee. Among size 1 firms, the three most capital intensive sectors are paper and
furniture, food and beverages and chemicals. The least capital- intensive sectors are textiles
and electronics and electrical machinery. Among size 2 firms, the two most capital- intensive
sectors are chemicals and food and beverages. The least capital- intensive sectors are textiles
                                                  -56-


and vehicles and auto parts. In size 3 firms, the most capital- intensive sector is metals and
machinery followed by textiles, paper and furniture and food and beverages. The least
capital- intensive sectors are vehicles and iron and steel. Across all size-classes, the relatively
capital- intensive sectors are paper and furniture and food and beverages.

    Table A3.2: Fixed capital stock and sales per employee, 1998
                          Capital stock in R ‘000 per        Sales in R ‘000 per
                                   employee                      employee
                          Size 1     Size 2    Size 3    Size 1     Size 2    Size 3
Chemicals                  195       333       98         372       535        579
Electronics & elect.       53        58        152        341       325        440
Mach.
Food & beverages           284       275       194        255       345        536
Iron & steel               108       126       97         334       254        285
Metal                      92        116       268        216       314        412
Products
Paper & furniture          263       130       208        255       333        362
Textiles                   25        41        260        109       174        167
Vehicles and auto parts    75        47        67         447       332        279

       Table A3.2 also reports sales per employee in 1998. Among size 1 and 2 firms,
vehicles and auto parts, electronics and electrical machinery and textiles have higher sales
per employee but low capital stocks per employee, suggesting a relatively high efficiency of
capital use. In size 3 firms, capital seems to be most efficiently used in the chemicals and
vehicles producing sectors followed by iron and steel, electronics and electrical machinery
and food and beverages. Across size-classes, vehicles and auto parts and electronics and
electrical machinery sectors seem to use fixed capital more efficiently than other sectors.
Distribution of firms by sector

The distribution of large manufacturing firms by sector in GJMA is quite similar to that in
the national economy, with a few exceptions. Table A3.3 show distribution of these firms in
Western Cape, Eastern Cape, Gauteng and Kwazulu Natal that collectively account for
approximately 85 percent of all large firms in South Africa. The first column under each size-
class, labeled SA, represents the sectoral distribution of firms in these four provinces of
South Africa. The second column in each size-class shows the sectoral distribution of firms
in the GJMA area that accounts for approximately 38 percent of SA firms. For example, in
the metal products sector nationally, size 1 firms represent 57 percent of all firms in SA,
while in GJMA this group represents 56 percent of all firms. In electronics and electrical
machinery, paper and furniture, vehicles and parts, and textiles, size 1 is dominant in GJMA
with over 40 percent shares across size-classes.

      Mapping the location of firms revealed that economic activity is concentrated across
the center of the city and stretching toward the northern and eastern industrial belts. There is
a distinct absence of industrial activity in the area south of the Central Business District
(CBD). This absence is important given that the bulk of GJMA’s labor force resides in the
southern areas including Soweto. As such, there is a disconnect between the factor markets
(labor) in the south and the location of production in the central, north, and eastern parts of
Greater Johannesburg.
                                                         -57-



Table A3.3: Sectoral distribution of firms in South Africa 49 and GJMA by size -classes
in 1998 (percent)

                                                 Size 1                 Size 2              Size 3
                                                 SA             GJMA    SA       GJMA       SA        GJMA
Chemicals                                        42             39      29       30         29        31
Electronics and Electrical machinery             45             48      26       26         29        26
Food and Beverages                               38             41      27       20         35        40
Iron and steel                                   33             41      26       23         40        36
Metals and products                              57             56      24       23         19        21
Paper and furniture                              42             51      28       25         30        24
Textiles                                         32             43      28       26         39        30
Vehicles and parts                               39             44      26       23         35        33
National distribution of young or post-apartheid firms

The grouping of young or post-apartheid firms that started production after 1995 contains
approximately 1000 firms that mostly fall into size-classes 1 and 2 (Chart A3.2).
Geographically, young firms are widely dispersed across GJMA with an apparent absence
just north of Johannesburg’s CBD.

Chart A3.2 Sectoral distribution of young (post-aparthe id firms), 1999



                                                                     Chemicals
                                Auto and parts
                                                                       20%
                                     4%

                Textiles
                 23%                                                                  Electronics & elec.
                                                                                            Mach.
                                                                                             18%

            Paper & furniture
                 11%


                           Metal products                                Food & beverages
                                                 Iron & steel
                                8%                                             14%
                                                     2%




49
  Excludes firms in the Northern Cape, Northern Province, NorthWest, Mpumalanga and Free State that
collectively account for 15 percent of South Africa’s large manufacturing firms.
                                                                                    -58-


Chart A3.3 Legal status of firms, 1999


                       100
                        90
                        80
    Percent of firms




                        70
                        60
                        50
                        40
                        30
                        20
                        10
                         0
                                               Size 1                        Size 2                              Size 3

                            PTY Ltd                            Closed Corporation                    Sole proprietorship
                            Subsidiary                         Partnership                           Privately held corporation
                            Co-operative



Chart A3.4 Ownership structure and percent of firm owned by type of investor, 1999


           60                                                                                                                                   100
           40
                                                                                                                                                50
           20
                       0                                                                                                                        0
                                Resident of SA       Domestic company Local bank.inst.investor         Non-resident        Foreign company



                                                   Percent of firms           Mean % share


Inventory, unsold output

LFS data indicates that large South African firms in manufacturing seemed to manage weak
market demand through efficient inventory management. On average, firms were able to sell
over 90 percent of the output they produced.
Chart A3.5 Percent of unsold output by sector, 1998


                             Auto and parts
                                    Textiles
                           Paper & furniture
                             Metal products
                                Iron & steel
                       Food & beverages
Electronics & elec. Mach.
                                 Chemicals

                                               0              2               4                  6                    8            10                  12

                                                                                                                                    Percent of firms
                           Percent of firms




                                                100
                                                           120




                  0
                      20
                           40
                                 60
                                       80
Crime and theft


    Cost of
  capital/credit

Depreciation of
    Rand

 Recent labour
  regulations

   Corruption in
   government
                                              SIZE 1

   Availability of
   labour skills
                                                                                      (Percent of firms)




     Tax rates




Crime and theft
                                                                 Major




    Cost of
  capital/credit

Depreciation of
    Rand

 Recent labour
                                                                                                                                                                                      -59-




  regulations
                                              SIZE 2




   Corruption in
   government

   Availability of
   labour skills
                                                                 Moderate




      Tax rates
                                                                            Chart A4.1 Firm rating of economic constraints by size -class




Crime and theft


    Cost of
  capital/credit

Depreciation of
    Rand

 Recent labour
  regulations
                                                  SIZE 3




   Corruption in
   government

   Availability of
   labour skills
                                                                                                                                            ANNEX 4. CONSTRAINTS TO BUSINESS GROWTH




     Tax rates
                                         -60-


     ANNEX 5. CRIME AND INFRASTRUCTURE IN GJMA

Map A5.1: Firms reporting low proximity to rail transport.




Map A5.2: Firms reporting low reliability of telecom services
                                                              -61-


Location of Input and Output Markets


Chart A5.1 Location of input and output markets, 1999

          Rest of World


   Rest of South Africa


               Gauteng

                  GJMA


                          0       10       20       30   40      50   60       70       80      90

            Inputs    Outputs                                              Percent of firms



Note: A firm can purchase inputs from multiple sources




Has service delivery improved in South Africa in the last three years?

Table A5.1 Improvement in ratings* for government provision of services

                                                         Today                                3 yrs ago
Argentina                                            2.61                              2.51
Brazil                                               2.23                              2.18
Chile                                                3.36                              3.24
Mexico                                               2.95                              2.64
UK                                                   3.40                              3.48
Germany                                              2.81                              3.01
Spain                                                3.87                              3.52
Italy                                                2.95                              2.63
United States                                        3.48                              3.29
Argentina                                            2.61                              2.51
Thailand                                             3.28                              3.93
South Africa                                         3.47                              3.59
*A rating of 1 means very inefficient and a rating of 6 means very efficient. WBES, 2000
                                                          -62-



           ANNEX 6. EXCHANGE RATE AND TRADE POLICY
Table A6.1 Proportion of trade by size -class

                     Percent of output                    Percent of inputs
                         exported                             imported
Size 1                     13.28                                35.09
Size 2                     17.98                                33.24
Size 3                     13.03                                35.85

Table A6.2 Percent of output exported and imported by sector

                                                    Mean percent of output    Mean percent of
                                                          exported            output imported
     Food/Beverages                                         14.32                  17.70
     Electronics & elec. machinery                          19.46                  32.83
     Iron/Steel                                             22.68                  22.49
     Metal Products                                         18.27                  22.26
     Vehicles                                               27.21                  35.75
     Chemicals                                              11.95                  48.84
     Paper/Furniture                                         8.03                  31.22
     Textiles                                                9.09                  45.34

Hedged firms response to the 1998 Rand crisis – comparing sizes 2&3 with size 1

Since only 25 percent of size 1 firms were hedged compared to 58 percent of size 3 firms,
they are examined separately to examine their response to 1998 Rand crisis. Chart 6.1 shows
the response of size 2 and 3 firms that were 100 percent hedged as compared to firms that
were not hedged. In these two size-classes, the dominant firm response was to “start
exporting more.” Almost 67 percent of the firms that were hedged 100 percent started
exporting more compared to only 46 percent of those that were not hedged. 50 In the case of
import substitution, the difference between those that were hedged and not hedged was small.

      Since size 1 firms were the least hedged, their behavior differed from those of size 2
and 3 firms and is illustrated on Chart A6.2. While it appears that on average, those who
were 100 percent hedged exported more, engaged less in import substitution, and were also
less uncertain of the future, these differences are not statistically significant.




50
     This result is significant at the 5 percent level.
                                                            -63-


    Chart A 6.1 Response to 1998 Rand crisis- Comparing firm sizes 2 & 3 which are
    hedged with those that are not hedged at all


Don't hedge-Uncertain of future: will wait and see

     Hedge -Uncertain of future: will wait and see

             Don't hedge Started exporting more

                   Hedge Started exporting more

            Don't hedge -More import substitution

                 Hedge -More import substitution

                                                      0           20   40           60      80        100
    Very much              Moderately                Not at all
                                                                                          Percent of firms




    Chart A 6.2 Response to 1998 Rand crisis- Comparing firm size 1 which are
    hedged with those that are not hedged at all

Don't hedge-Uncertain of future: will wait and see

    Hedge -Uncertain of future: will wait and see

             Don't hedge Started exporting more

                   Hedge Started exporting more

           Don't hedge -More import substitution

                 Hedge -More import substitution

                                                     0         20      40            60          80     100
                            Very much                Moderately        Not at all
                                                                                          Percent of firms
                                                  -64-


             ANNEX 7. EMPLOYMENT PROFILE IN GJMA

Sectoral distribution of employment

Chart A7.1 Sectoral distribution of jobs in GJMA, 1998

                           Vehicles
                             4%
                                                   Chemical
        Textiles                                     17%
          7%



  Paper Furniture
                                                        Electrical
      10%
                                                          15%

     Metal
     16%
                 Iron
                 3%
                                          Food
                                          28%




Chart A7.2 Sectoral distribution of jobs in South Africa, 1998


                               Vehicles
                                 4%         Chemical
                                              11%
        Textiles                                   Electrical
         16%                                          9%



      Paper
     Furniture
       12%


     Metal
      9%                                         Food
                                                 37%
                        Iron
                        2%
                                                       -65-


Gender profile of large firm employees

Chart A7.3 Male-female shares in occupational categories in GJMA


                                           Males     Females
       90
       80
       70
       60
       50
       40
       30
       20
       10
        0
             Managerial    Prof &     Clerical     Service     Craft   Plant op   Laborers
                          Technical




Capital per worker by firm size

A measure of the labor- intensity of firms is the employee to capital assets ratio. Several
factors complicate a precise measurement of this ratio from the LFS data. First, the survey
year, 1999, coincided with a period of fairly low economic activity. Second, the survey year
coincided with a period when there were large-scale layoffs in the manufacturing sector.
Third, since the early 1990s, firms have undergone significant rationalization that has
prompted use of new production techniques and adoption of new technology. In the midst of
these ongoing structural changes, it is difficult to gauge precisely whether the observed labor-
intensity is indeed reflective of the firms’ choice of production techniques. Nevertheless, the
labor intensity of firms in the manufacturing sectors was measured from LFS data and is
reported below.

      For every 100,000 rands of existing capital in a firm, the LFS registers an average 51 of
1.7 employees. We found that firms across the three size-classes do not display statistically
different labor- intensity. 52 The LFS found that across sectors in GJMA, firms have roughly
the same labor-intensity per R 100,000. The exception is the textiles sector which, on
average, employs 4.6 or about twice as many workers than the next most labor intensive
sector, metal products, which employs only 2.3 workers per R 100,000 (Table A7.1). Food
and beverages is the least labor-intensive with a ratio of 1 per R100, 000.




51
   The median is 1.7 and the mean is 5.5. Since there are large outliers (the max is 82 employees per R 100,000
in one firm), the median is a better measure.
52
   All results are significant at the 5 percent level.
                                                          -66-




Table A7.1 Labor-intensity across sectors
                                                           Employees/ R 100,000 in
                                                                  capital
     Chemicals                                                      1.4
     Electronics and electrical machinery                           2.2
     Food and beverages                                             1.0
     Iron & steel                                                   1.5
     Metal products and machinery                                   2.4
     Paper & furniture                                              1.4
     Textiles                                                       4.6
     Vehicles & auto parts                                          2.1
     ALL                                                            1.7

       By size-class, size 1 and 2 firms differ from size 3 firms 53 in the intensity with which
they employ less-skilled workers per R 100,000 of existing capital. While size 3 firms
employ only 0.25 laborers, size 1 and 2 firms employ as many as 0.5 to 0.6. Across other
skill-types, the labor- intensity of production is statistically similar, in spite of the graphical
differences across firms, as for example, in the case of semi-skilled workers.




53
     This result is significant at the 5 percent level.
                                                           -67-



                         ANNEX 8. LABOR REGULATIONS

Chart A8.1 Effect of bargaining council system on employment in GJMA


  Size 3


  Size 2



  Size 1


           0       10     20        30      40       50        60      70       80           90        100

           Increases     Declines        No effect        Not applicable             Percent of firms




Effect of individual labor regulations on employment in GJMA

Chart A8.2 Effect of Labor Relations Act on employment in GJMA


  Size 3


  Size 2


  Size 1


           0       10     20        30      40       50        60      70       80           90        100
                                                                                  Percent of firms


            Raise employment    Lower employment          No effect   Not familiar with it        NA
                                                          -68-


Chart A8.3 Effect of Basic Conditions of Employment Act on employment in GJMA



  Size 3



  Size 2



  Size 1


           0      10      20       30      40      50         60      70        80         90        100
                                                                                 Percent of firms


     Raise employment          Lower employment    No effect        Not familiar with it        NA

Chart A8.4 Effect of Employment Equity Act on employment in GJMA


  Size 3


  Size 2


  Size 1


           0      10      20       30      40      50         60      70        80         90        100
                                                                                  Percent of firms


           Raise employment     Lower employment        No effect   Not familiar with it        NA

Chart A8.5 Effect of Skills Development Act on employment in GJMA



  Size 3



  Size 2



  Size 1


           0      10      20       30      40      50         60      70        80         90        100

                                                                                     Percent of firms

      Raise employment         Lower employment     No effect       Not familiar with it        NA
                                                     -69-




                         ANNEX 9. AVAILABILITY OF SKILLS
  Table A9.1: 1998-99 Vacancy Rates

                                         0%             Less than 20 %     21- 50 %     More than 50 %
Firm Size 1 (50 – 99 workers)     1998    1999              1998   1999   1998   1999    1998    1999
Managerial                        79.9        86.0          9.9    6.8    8.9     7.1     1.2     0
Professional & technical          87.6        90.6          5.4    5.3    6.9     2.5     0      1.6
Clerical                          91.3        95.7          4.6    0.2    4.2     4.0     0       0
Service                           88.5        90.1          1.8    2.6    8.1     5.0     1.6    1.6
Craft & related trades            85.3        87.9          9.5    8.5    3.4     3.6     1.7     0
Plant operators                   93.0        89.1          4.2    6.7    1.4     4.2     1.4     0
Laborers                          91.5        93.8          6.9    6.1    1.4     0       0.1     0

Firm Size 2 (100 – 199 workers)   1998    1999              1998   1999   1998   1999    1998    1999
Managerial                        86.9        84.9          13.1   6.8     0      7.1     0       0
Professional & technical          82.1        80.1          12.7   5.3    5.1     2.5     0       0
Clerical                          96.5        95.3          1.9    0.3     0      4.0     1.6     0
Service                           80.9        80.6          6.9    2.6    9.3     5.0     3.2    4.7
Craft & related trades            90.4        89.3          3.7    8.5    3.9     3.6     1.9     0
Plant operators                   94.0        93.7          5.9    6.7     0      4.2     0       0
Laborers                          95.1        95.3          4.9    6.1     0      0       0       0

Firm Size 3 (200 plus workers)    1998    1999              1998   1999   1998   1999    1998    1999
Managerial                        74.2        70.2          21.5   29.8   4.3     0       0       0
Professional & technical          76.1        68.6          23.8   24.3    0      7.0     0       0
Clerical                          77.2        77.5          19.8   18.0   2.9     4.5     0       0
Service                           84.3        77.1          15.7   21.1    0      1.8     0       0
Craft & related trades            88.8        83.6          10.6   15.3   0.6     0.3     0      0.7
Plant operators                   92.7        79.9          7.3    20.8    0      0       0       0
Laborers                          96.0        97.8          3.9    0.3     0      1.9     0       0
                                                         -70-


Recruitment preferences of large firms

Chart A9.1 Recruitment pre ference for production workers in GJMA by age


  46 and above



        25-45



        16-24


                  0    10      20       30      40     50       60   70      80       90    100
                                                Percent of firms
         Preferred     Not Preferred




Chart A9.2 Recruitment preference for production workers in GJMA by gender




  Not Preferred




     Preferred




                  0     10     20       30      40     50       60   70       80      90     100
            Male      Female   Either
                                                                              Percent of firms

Chart A9.3 Recruitment preference for production workers in GJMA by work
experience




                                Not Preferred
                                    25%



                                                                          Preferred
                                                                            75%
                                                                       -71-




Chart A9.4 Recruitment preference for production workers in GJMA by education



  Not Preferred




      Preferred



                  0          10        20         30        40        50        60    70      80      90       100

     Primary          Secondary             Vocational           Tertiary                   Percent of firms




Chart A9.5 Recruitment preference for production workers in GJMA by work
experience and education


  Work experience +
       tertiary

  Work experience +
     vocational

  Work experience +
     secondary

  Work experience +
       primary

                         0        10         20        30        40        50    60    70      80     90       100
     Preferred        Not Preferred         N/A                                             Percent of firms
                                                                              -72-


               Incidence and magnitude of training by sector
                In-house training

               Percent of firms that offer in-house training
                                         Chemical Electrical      Food        Iron        Metal        Paper/furn    Textiles     Vehicles     All
               Managers/Proefessionals      53         33         26          37           34             30           36           20         36
                    Semi-skilled            61         46         42          58           45             42           23           55         47
                     Unskilled              46         33         39          53           39             37           13           49         38
                   Common skills            18         16         23          29           15             16           13           24         17

Average number of training programmes offered in-house
                                         Chemical Electrical      Food        Iron        Metal        Paper/furn    Textiles     Vehicles     All
               Managers/Proefessionals      6.1        2.8        12.3        4.7          3.7            5.4          1.9         11.2        5.2
                    Semi-skilled            8.9        3.6         4.7        6.4          5.6            2.3          1.3          3.4        5.4
                     Unskilled              5.0        1.9         2.9        4.2          3.8            0.8          0.0          1.4        3.1
                   Common skills           12.5        3.7         6.7        4.0          1.6            1.8          1.0          6.9        5.4

      Median number of workers trained in-house
                                         Chemical Electrical      Food        Iron        Metal        Paper/furn    Textiles     Vehicles     All
               Managers/Proefessionals      19         20         40          9            10             12            6            30        9
                    Semi-skilled            30         21         20          17           20             18            9            30        16
                     Unskilled              23         26         40          30           13             21           30            32        20
                   Common skills            75         60         598         41           40             40           75           100        51

                Outside training

               Percentof firms that offer outside training
                                         Chemical Electrical      Food        Iron        Metal        Paper/furn    Textiles     Vehicles     All
               Managers/Proefessionals            38         54          48          46           47            20           38           37         42
               Semi-skilled                       35         41          18          46           48            24           38           40         37
               Unskilled                          11          6          11          19            8            13            0            8          9
               Common skills                      10          8          11          16            7             7            0           12          8

               Average number of outside training programmes offered
                                         Chemical Electrical      Food        Iron        Metal        Paper/furn    Textiles     Vehicles     All
               Managers/Proefessionals      9.2        5.0        4.4          4.5         4.2            3.7          2.6          4.7        5.3
               Semi-skilled                15.6        4.0        4.4          3.4         2.9           12.2          4.9          6.8        6.9
               Unskilled                    1.3        2.1        0.6         41.1         2.5            0.8          N/A          1.1        3.8
               Common skills               20.4        6.5        2.8          4.5         4.7            1.4          N/A          8.0        8.9

               Median number of workers trained outside
                                         Chemical Electrical      Food        Iron        Metal        Paper/furn    Textiles     Vehicles     All
               Managers/Proefessionals      10         6          11          8             4             11           13           8          8
               Semi-skilled                 13         5          40          6             8             11           15           14         8
               Unskilled                    22         30         60          22           18              4           N/A          25         20
               Common skills                40         15         92          20           44              6           N/A          45         28


               Percent of workers trained among those who do any training at all - in-house plus outside
                                         Chemical Electrical      Food        Iron        Metal        Paper/furn    Textiles     Vehicles     All
                 Percent of workers in
                 firm who were trained      65         20         35          35           30             24           11           41         33
                                           -73-
Box A9.1: From Providing Training to Providing Information

Retraining is vital for a country’s economic restructuring. But mismatches between formal training programs and
the evolving skill needs of employers have been a persistent problem, often resulting in very low rates of return to
resources devoted to such training. One way to improve performance is to move from direct provision to
government intermediation between provider and trainee. This can reduce the information costs that lead to such
mismatching.

        A program proposed in Madagascar will serve workers in 45 state enterprises facing restructuring, in
preparation for their privatization or liquidation. A previous failed attempt to facilitate these workers’
redeployment was heavily centralized and information-intensive. The agency in charge had to identify the
sectors and activities in which the soon-to-be-separated workers could work and then provide them with
appropriate training. And it had to help those planning to launch their own micro-enterprises to design a business
plan and buy the appropriate equipment. Many separated workers received their equipment years after losing
their jobs. By that time they had moved on to other activities, so most of them simply sold the equipment.
Despite costs of roughly $900 per worker, dissatisfaction with the program was widespread.

       In the new redeployment program, the agency in charge provides a menu of training and redeployment
services, but separated workers choose whether to “buy” those services. First, from a study of labor market, the
agency calculates the present value of the earnings loss each separated worker will experience. Simultaneously,
the agency launches a tender for redeployment services, which is open to other government bodies, private
providers, and non-government organizations. Next, the agency runs a w      orkshop at the plant, where all the
retained bidders describe to the workers to be separated the redeployment service they propose. Finally, each
separated worker decides which services to buy.

       The government agency discounts the cost of these services from the amount of assistance allocated to
each worker, pays the rest in cash, and pays the retained bidders upon delivery of their services. Workers have
the right to get all their compensation in cash if they do not consider the services worth their cost. This
minimizes the risk that large sums will be wasted in useless training and redeployment efforts.
Source: World Development Report 1998/99.




Box A9.2 Enterprise Training in Malaysia in 1995
                                                          Small*       Medium         Large
Investment in training (percent of firms)
All formal training                                         18.2          44.7         70.7
In house                                                    13.5          31.7         53.6
External                                                    7.6            27          51.4
Top training providers (percent of firms
who train)
Private training institutes                                 28.6          44.3          53
Buyers/Material suppliers                                   24.5          25.1          25
Other government institutes                                 20.4          22.7         27.1
Skills development centers                                  10.2          14.9         28.8
Joint venture partners                                      10.2          9.8          11.9
Overseas training                                           8.2           12.9         21.2
Number of workers trained (as percent of
firm labor force)
All forma l training                                        10.4          13.2         29.5
In house                                                    9.2           10.5         26.1
External                                                    1.1           2.8          3.4
*Small: 16-100 workers, Medium: 101-250 workers, Large: over 250 workers.
Source: Malaysia: Enterprise Training, Technology and Productivity, The World Bank, 1997.
                                                     -74-


The Malaysian experience with firm training

A recent World Bank Study54 evaluated two training programs in Malaysia. The first
program is the Double Deduction Incentive for Training (DDIT). This program permits
employers to deduct double the amount of allowable training expenses on their tax returns.
Employers could send their workers to approved training institutions or apply directly to the
industrial development authority for approval of their planned training programs. The report
found that Malaysian firms that do little or no training had extremely low awareness of the
DDIT program. Use of the DDIT program is dominated by MNC’s. Only 7 percent of
applications to the industrial board were from small firms and over 40 percent were from
firms with over 500 employees. Moreover, since the DDIT program was used primarily be
firms that were already doing training, it has translated into windfall gains for large
companies and loss of tax revenue for the Government.

    The second training program evaluated is the Human Resource Development Fund
(HRDF) which is administered by a council of representatives from the private sector and
government. Employers who have contributed a minimum of six months are eligible to
claim a portion of allowable training expenditures up to the limit of their total levy 9 (one
percent of payroll) for any given year. The HRDC consists of a number of different and
flexible training schemes, some of which are directed at small and medium enterprises,
which include:

•     Employers can send employees to approved training courses offered by registered
      providers and submit claims on completion (ATP Scheme).
•     Employers can submit plans for approval of training courses offered by non-registered
      providers (SBL Scheme).
•     Employers can submit detailed annual training plans that cover at least 10 percent of the
      firm’s workforce (PLT Scheme).
•     Training providers are enlisted as agents to the firm in order to lower firm’s training cash
      outlays (PERLA Scheme).
•     Pre-approved training programs are given a special designation allowing training
      providers to market such training and employers to simplify their reimbursement claims
      (SBL Pre-Approved Scheme).
•     Training needs analysis workshops are provided to answer questions about different
      schemes, provide assistance in the purchasing of training aids and training rooms (TNA).
•     Group training of small and medium enterprises is provided (Joint Training Scheme).
•     Training to employer associations to play a greater role in developing training programs
      is provided (Group Training Scheme).

    The HRDF was found to have successfully provided firms with alternative schemes for
training both in- house and external. The program’s success is related to its streamlining of
reimbursements and expenditures, its reduction of administrative and reporting burdens on
employers, and its wide disseminated information. However, the program is constrained in
that non-compliance is high.


54
     Malaysia: Enterprise Training, Technology, and Productivity. The World Bank, Washington, DC, 1997.
                                                      -75-


   Based on analysis of these two programs, a number of policy options arose. Some of the
most pertinent points are discussed below:
1. Collection and Dissemination of Training Information: Training information needs to be
   collected more systematically and needs to be made more available to individuals and the
   private sector through periodic, timely publications. Information needs to be gathered for
   monitoring purposes. The report recommends that the Government sponsor enterprise
   and household surveys on periodic basis and that an inter-agency steering committee
   designs and coordinates the surveys.
2. Expanded Role of Training Institutions: The government should select public training
   institutes for corporatization which would compete with private training sources and cater
   to needs of industry. Skills Development Centers have been a productive instrument
   provided by government tho ugh their ability to attract small and medium enterprises
   needs to be addressed. Government should also examine the possibility for bilateral
   training institutes with OECD that have proven to work well.
3. More Effective Training Policies: Information about policy initiatives needs to be widely
   publicized. Most fail due to poor awareness. TV and newspaper campaigns should be
   mounted to encourage firms to sign up. Filing requirements for incentives should be
   streamlined and less costly and implementation needs to be closely monitored.

      With regard to payroll contributions of smaller firms, firms should register in the
incentive schemes and contribute at a reduced rate with the government matching firm’s
contributions. The government contribution should be time limited and fall over time as
small firms’ gain training capabilities. Programs aimed at small firms should be coordinated
with small business development centers.

Box A9.3: Malaysia: Public-Private Collaboration in Training Provision

The Penang Skills Development Center (PSDC) in Malaysia demonstrates how public-private cooperation can be
harnessed to provide demand-driven training programs for regionally based industry. Established in 1989 as a tax-
exempt institution on the joint initiative of the Penang state government and major manufacturing firms, the PSDC
is funded by corporate contributions and training fees, and by the state government which provides PSDC with
subsidized facilities. Its strategic location in the Bayan Lepas Free Trade Zone provides worke rs with easy access to
training, which they attend full-time or on a part-time basis. Training programs and course content are determined
by a training committee on the basis of semi-annual needs assessments of firms and approved by the PSDC council
representing the state government, corporate members, the local university, and the national standards agency.
Training courses are taught by staff from external training institutions, the local university, and the firms
themselves. Firms have several incentives to join and work with PSDC: they help shape the training agenda; they
pay relatively low training fees because of economies of scale; they have access to classrooms and recent vintage
training equipment; equipment donations to PSDC are tax-deductible and, as donors, they receive preferential access
the equipment for running their own training courses.

The Malaysian government has sought to replicate the PSDC model in other states by providing grants to
complement contributions of land and training facilities by state governments. Early experience with these SDCs
has been mixed. The most dynamic SDCs had several characteristics --- they brought private sector firms into the
planning process from the very beginning; training need assessments were conducted of firms in the region to be
served; machinery suppliers were convinced to donate equipment for training; and training facilities were located
close to firms. Some SDCs were less successful, in some cases because of poor location of training centers (location
decisions were shaped by a policy of decentralizing industry from congested areas), in other cases because the
training provided was supply-driven (planning focused more on assessments of the training capabilities of existing
providers than on the training needs of firms).
Source: New Foundations – Private Sector Development in Post-War Bosnia and Herzegovina, World Bank, 1997
                                                                -76-


Box A9.4: Enterprise Training and Productivity; Main Findings
The Private Sector Development Department (PSD) of the World Bank conducted a widespread survey in
1993 on enterprise training in Indonesia, Malaysia, Colombia, Mexico and Taiwan. Their main findings from
this survey include:
• Firms under invest in training. Between 50 to 80 percent of small firms and 20 to 70 percent of larger
     firms do not give their employees formal structured training. Market failures are an important constraint to
     training. Although the majority of firms attribute lack of training to mature technology, a large number of
     firms have problems with free riding (i.e. trained labor leaves for higher wages at another firm.), lack of
     knowledge about training methods, and/or limited resources.
• Training raises firm level productivity. PSD analysis of the survey data found that firms that train are
     about 32 percent more productive than firms that don’t train. Productivity effects are strongest in
     Indonesia where firms that train are 71 percent more productive and in Mexico where firms that train are
     44 percent more productive. In general, productivity effects are strongest for smaller firms, exporting
     firms, and firms that train externally.
• Private training sources are more important than public ones. Between 21-31 percent of firms in
     Malaysia and Colombia used either private training institutes, suppliers, or industry associations for
     external training needs. 13 percent of firms in Indonesia used private training sources. When PSD
     analyzed the efficiency of government provided training, the team found that in both pre-employment and
     post-employment training, public training institutions did not increase firm productivity.
• Technology shapes skill requirements. Firms are more likely to train when they are large and employ an
     educated work force, invest in R&D, possess technology, have foreign capital participation, use quality
     control methods and export.
• Firms that train pay higher wages. The wage increase from training is on average 6 percent and mirrors
     productivity gains. As such, productivity differentials from training may lead to wider wage disparities
     between skilled and unskilled labor.
•    Compensation benefits deter quits: Policies such as pay increased tied to length of service, retirement
     schemes, benefits etc and severance pay were effective.
Source: Enterprise Training in Developing Countries: Incidence, Productivity Effects and Policy Implications, World Bank, 1995.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:18
posted:2/16/2011
language:English
pages:87