Protecting Workers, Creating Jobs: Rethinking Social Protection in Developing Countries

1
Protecting Workers, Creating Jobs Rethinking
Social Protection in Developing Countries

• Carmen Pagés
• Social Protection and Labor Division
• The World Bank

2
This talk

• Market economies generate high unemployment risk
• Facing up to unemployment riskWhat is there now?
• Improving income security How to go about it?

3
Substantial labor market risk All countries
experience very large rates of job turnover
Average net
Gross Job Flows (average)
Change in employment
( Employment)
Mexico (1994-2000)
6.0
Brazil (1991-2000)
1.1
Denmark (1983-1989)
2.2
-0.1
New Zealand (1987-1992)
Sweden (1985-1992)
-4.1
Canada (1983-1991)
2.6
France (1984-1992)
0.7
Italy (1984-1992)
1.2
U.S. (1979-1983)
1.5
Finland (1986-1991)
-1.6
Estonia (1992-1994)
-3.2
1.5
Germany (1983-1990)
-25
-20
-15
-10
-5
0
5
10
15
20
25
Entry
Gross Job Destruction
Exit
Gross Job Creation
Source
OECD (1996), Davis Haltiwanger (1999),
Kaplan, Gonzalez Roberts (2003) and
Menezes-Filho (2003)
.
4
Substantial labor market-generated risk

• Importance of firm entry and exit
• Larger degree of job creation and destruction in
  small and young firms and in unskilled
  labor-intensive sectors.
• Importance of idiosyncratic shocks
• Simultaneous job creation and destruction both in
  recessions and expansions.

5
Job market related risks are even higher in
developing countries

• Larger proportion of small firms and unskilled
  labor-intensive products.
• Larger and more frequent macroeconomic shocks.
• Large prevalence of unregulated, unregistered jobs

6
The probability of entering into unemployment is
higher for informal than for formal workers
Source Duryea et al (2006)
7
Yet, mounting evidence suggest that job
reallocation is key for productivity growth
Within firm productivity growth
Reallocation across firms
Firm Entry
Firm Exit
Productivity Growth
Colombia 1992-97
Chile 1992-97
U.S 1992-97
1987-92
UK 1992-97
1987-92
Portugal 1992-1997
1987-1992
Holland 1992-97
1987-92
Italy 1992-97
1987-92
Germany 1992-97
France 1987-92
Finland 1989-94
1987-92
-75
-25
25
75
125
175
Source
Scarpetta, Hemmings, Tressel y Woo (2002),
Medina, Melendez y Seim (2003) and IADB (2004).
8
Process of Growth ? Creates risks for workers

• Markets alone cannot diversify this risk No
  country has been able to develop a private market
  for unemployment insurance (UI)
• Without proper insurance some of this growth may
  not take place as workers are not willing to
  accept growth enhancing changes (Boyer, 2006)
• Insufficient UI leads to inefficient job matching
  (Acemoglu and Shimmer, 1999)

9
Insecurity Unemployment is the main source of
concern for Latin Americans
Source Latinobarometer, 2003
Most pressing problem in your country
Unemployment
Corruption
Poverty
Delinquence /Insecurity
Employment Instability
Low wages
Terrorism/Political Violence
Low quality of Education
Inflation
Lack of opportunities for youth
Drug Consumption
Human Rights Violations
Health Problems
Drug traffic
Housing problems
Transport
Environment
Racial Discrimination
0
5
10
15
20
25
10
Focusing on developing countries What
institutional arrangements are in place to face
up to these risks?
Typical labor market institutional set up in many
developing countries (based on Latin America
South Asia)
High levels of (de Jure) Job Security
Large Informal Sector where regulations are not
enforced or do not apply
Little or no coverage of unemployment
insurance
Hardly any (effective) presence of
intermediation and employment services
State provided VET Lack of incentives for
on-the-job training. Little re-training for U
Low prevalence and coverage of collective
Bargaining
Low institutional state capacity Low enforcement

Some public works programs in crisis periods
11
Employment protection is higher in developing
countries
Source Doing Business Indicators, World Bank
2004
12
Fragmented, insufficient ineffective welfare
state.

• Large informal sector ? many workers are not
  likely to be protected by EPL in the likely case
  they become U.
• Unemployed workers have little help in finding
  another job. Placing rates in employment
  services are low geared to skilled workers,
  public employees (Okeefe 2006, Mazza, 2002)
• Some vocational training (often outdated) but
  little on the job training or re-training for the
  unemployed
• Pension and health programs for formal
  salaried---but low coverage density of
  contributions (Rofman, 2005 Bertein, Larrain
  Pino, 2006)

13
Fragmented, insufficient ineffective welfare
state.

• Need to develop a system that covers all workers
  (with a combination of different programs)
• Income support for formal workers
• EPL, UI, UISA?
• Income support for informal workers
• Unemployment Assistance, Individual Contribution
  Programs, Public works?

14
Income support (IS) for formal workers EPL or UI?

• EPL
• It protects the job reducing job flows
• It requires little from state (only enforcement)
• Firms pay for the unemployment they generate
• Firms have to pay when they undergo financial
  problems, bankrupt firms dont pay.
• It does not protect against U duration risk
• It pools risk within the firm

• UI
• It protects the worker
• It is quite demanding institutionally
• May lead to excessive turnover unless UI is
  experience rated.
• State has to pay more in bad timesconstrained by
  pro-cyclical fiscal policy
• It does, but it may lead to excessive U duration
  if too generous.
• It pools risk across (formal) workers

15
Other problems associated with EPL

• Evidence from India (Besley and Burgess, 2004
  Ahmad Pagés, 2006) indicates that job security
  regulations
• Reduce creation of formal salaried jobs
• Increase the size of the informal sector
• Are not able to increase workers labor share
• Slow down investment and output growth (therefore
  workers get an equal share of a smaller pie)

16
Other problems associated with EPL

• Micco and Pagés (2006) find that employment
  protection reduces flows but also employment (and
  output) in sectors that require high flexibility
  (for example textiles).
• However, we also find that the effects dissipate
  in countries with less effective rule of law (due
  to lack of enforcement)

17
Other problems associated with EPL

• Employment protection reduces employment of
  young, women and unskilled workers relative to
  the adults, men and skilled workers (Khan,
  Bertola, and Blau, 2002 Montenegro and Pagés,
  2004)
• In OECD countries, EPL has been associated with
  higher U duration (Nickell, 1997). In developing
  countries, this effect is less important as
  workers find jobs in informal sector

18
Income support (IS) for formal workers UI or
UISA?

• UI
• It protects the worker
• It is quite demanding institutionally
• May lead to excessive turnover unless is
  experience rated.
• State has to pay more in bad timesconstrained by
  pro-cyclical fiscal policy
• It protects against unemployment duration risj
  but it may lead to excessive U duration if too
  generous.

• UISA
• It protects the worker
• It is less demanding institutionally (no moral
  hazard)
• May lead to excessive turnover unless
  contributions are experience rated.
• It does not pool risk neither across good and bad
  times, nor within firms or across workers.
• It does not protect against unemployment duration

19
No system is perfect some trade offs involved

• In countries with extremely weak state capacity,
  EPL may be the only way.
• Blanchard (2002) Characterizes EPL as
  Proto-Insurance
• Yet, system can be improved to
• Rely less on administrative barriers and more on
  severance pay (which benefit workers who end up
  unemployed).
• Mandating firms to held reserve accounts in
  financial institutions to which they have to
  contribute periodically.
• In countries with a more developed State,
  shifting from EPL to some sort of UI could avoid
  many of the pitfalls of EPL
• Chile new UI based on a combination of UISA and
  subsidies to unemployed workers with low deposits
  in accounts.
• Yet, need to evolve rather than superpose
  programs (unlike Chile)

20
Up to now labor reforms have been difficult, but

• Objective of most labor reforms has been to
  flexibilize.. Need to evolve to a mindset where
  objective of reforms is to improve income
  protection for workers.
• To improve upon the current scenario the current
  form of protection (i.e EPL) may need to be
  substituted for something else (it depends on the
  country)
• In many countries urgent need to simplify labor
  codes and improve inefficient (and often corrupt)
  enforcement mechanisms.

21
Income support (IS) for informal workers

• Special codes for informal firms workers need
  to be examined carefully as they may provide
  incentives for remaining small or improductive.
• In some countries, informal workers could be
  covered by assistential non-contributory
  mechanisms (zero pillar or unemployment
  assistance). Yet the factibility and design of
  such mechanisms needs to be studied carefully.
• Public work programs could be an important part
  of the equation if conceived as an UI mechanism
• Self-targeting (wagesltmarket wage or minimum
  wage)
• Expanded in contractions/geographic locations
  undergoing shocks
• Combined with worker training or job search
  assistance.

22
In conclusion

• Income support in developing countries is
  inadequate, inefficient and insufficient.
• Much can be done but need to get out of two
  opposite mind sets
• Flexibilization
• The status quo is OK (it may suit a few, but not
  adequate for all)
• Trade offs involved Initiatives should start
  small and be rigorously evaluated