Risk Management Strategies and Labour Productivity in Rural Africa

1
Risk Management Strategies and Labour
Productivity in Rural Africa

• Taro BOEL
• Food Security and Sustainable Development
  Division
• United Nations Economic Commission for Africa
• African Economic Conference
• 11-13 November 2009
• Addis Ababa

2
OUTLINE

• Context of risk management in Africa
• What are the risk management strategies?
• What link with labour productivity in Africa?
• How to improve labour productivity in rural
  Africa?

3
Context of risk management?

• The African farmer is facing various risks
• Four types of risks are identified by Weinberger
  and Jutting (2000)
• Health risks are among the most important in
  Africa (eg. De Weerdt and Dercon ,2006 in
  Kagera,Tanzania)

4
What are the risk management strategies?

• To reduce the exposure to shocks (risk management
  strategies) Ex ante actions
• To deal with the shocks (risk coping
  strategies)ex post actions

5
What link with labour productivity in rural
Africa?

• Risk Management Strategies
• Fluctuations of food consumption
• Health
• Weak capacity of performing physically
  demanding work
• Labour Productivity

6
The model and data

• Model
• varbmiit f (ageyrit, ageyrsqit, genderi,
  ruralit, grdit, illdaysit, explhlthit,
  varexpfoodit, hhsizeit, cleanwaterit, incagrit,
  incemplit,increntit, inctransit, savingit,
  acqsloanit, error term)
• Data used
• Panel data Kagera Health and Development Survey
  from 1991 to 1994
• LSMS type of survey with some modifications
  panel data, more detail information on some
  variables

7
Estimation of the model

• Ordinary Least Squares (OLS) Estimator not
  suitable because of unobservable factors (genetic
  factor)
• Choice between Random Effect and Fixed Effect
  Hausman test not very conclusive
• Random Effect to estimate the coefficient of
  time-invariant regressor such as gender

8
Econometric issues

• A Chicken-and-egg problem (endogeneity) Does
  good health (BMI) lead to higher income? Or does
  higher income lead to better health?
• Hausman Test The P-value of the test is 0.21 so
  I cannot reject the null hypothesis of exogeneity

9
Main Results

• Expenditures related to health have a negative
  effect on the variation BMI
• Income from transfers has a positive effect on
  the variation of BMI (same size)
• Income from employment has a negative effect on
  the variation of BMI.

10
Main Results

• All types of income are jointly significant while
  some (incagr and incrent) are not significant
  individually
• Sensitivity analysis Hausman Taylor Estimator,
  robust results for sign but not for size of
  coefficients and their significancy
• Why formal health insurance? Because when people
  are insured they do not wait until the illness
  gets worst.

11
How to improve labour productivity in rural
Africa?

• Policy implications
• International Community reinsurance or
  subsidies to help excluded poor people to have
  access to insurance schemes
• Public intervention to address weak links such
  as regulatory framework, its enforcement and
  quality of health care
• Private sector seize the opportunities in the
  demand of insurance.

12
Thanks for your kind attention