Labor demand shocks, unemployment, and suicide: evidence from provinces across Iran

ABSTRACT

In this paper, we investigate the relationship between the unemployment rate and suicide rate using data from all provinces of Iran and over the years 2009–2015. We exploit the variations in national industry-specific labor demand changes and province-level industry-composition as plausibly exogenous shocks to the province-level employment growth. Using a 2SLS-IV approach, we find that a 1 percentage point increase in the unemployment rate is associated with 0.72 and 0.68 higher incidences of suicide attempt and death per 100,000 population, respectively. Although both OLS and 2SLS-IV approach reveals consistent and statistically significant results, the marginal effects of the 2SLS-IV approach are larger, implying that Endogeneity issues underbias the OLS results of the relationship between unemployment rates and suicide rates.

KEYWORDS:

• Unemployment
• suicide
• panel data
• Iran
• mental health

JEL CODES:

• C26
• J17
• J23
• I15

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 These data are available in a series of PDF files. For an English version, refer to: https://www.amar.org.ir/english/Iran-Statistical-Yearbook

2 In 2011, the province of Tehran was divided into two smaller provinces: Tehran and Alborz. To have a consistent statistic for all variables, we aggregate the data for these two provinces for the years after 2011. Therefore, our final sample has 30 provinces for all years.

3 Note that Iraq is the western neighboring country of Iran.

4 Source: ssis.sci.org.ir.

5 We use ten broad categories of industries which are consistent across the Household Expenditure and Income Survey as well as the Census data. We use aggregated industry definitions. The parameter $k$ in Equation (3) represents 10 industry-level codes as follows: (1) agriculture, fishing, and forestry; (2) Mining; (3) Manufacturing; (4) Utility and Construction; (5) Trade; (6) Tourism and Transportation; (7) Services (financial, public administration, education, health, insurance, social work); (8) Private household production and services; (9) All other services; (10) Other unspecified-unknown-unclassified industries.

6 This is computed by the coefficient in column 4 of Table 3, the average suicide death in Table 1, and the fact that total suicide death in 2015 was 3,234 counts.