ABSTRACT
Governments have developed small loan programmes with a reduced interest rate to decrease unemployment in Iran. Using longitudinal, firm-level data from 2005 to 2010 in Iran, this study examines the effect of one Iranian province’s loan programme on employment based on two different methods of evaluating causal effects. The first method applies a difference-in-difference fixed effects matching estimator to estimate the employment effect of the programme. The second method applies the generalized propensity score to estimate the impact of the amount of a loan on employment. The results from the first method suggest that the loan programme has a positive and significant effect on the employment of treated firms. The results from the second method suggest that the estimated employment effects increase with the amount of the loan, whereas there is a decrease in the marginal effects of an additional amount of loan.
Disclosure statement
No potential conflict of interest was reported by the author(s).
ORCID
Javad Nosratabadi http://orcid.org/0000-0001-6680-6363
Notes
1 A province in Iran.
2 The exchange rate with the U.S. dollar was around $1 = 10000 IRR during programme years.
3 The interest rate of borrowing a loan from private banks was much higher than this range.
4 The author does not have access to information about the exact reasons for rejecting applications.
5 The author does not have access to the years 2010 and 2011 information. Nevertheless, this information is not used for estimating in this study and the main data set will be subsequently discussed.
6 All of these firms received a loan with the same interest rate (4%) and did not take up loans from other sources outside the programme during the programme years.
7 These firms might apply to a loan from the programme and were rejected and they did not take up loans from other sources outside the programme during the programme years as well.
8 This interest is subsequently explained.
9 For all details regarding this method see Flores et al. (Citation2012) or Bia, Flores, Flores-Lagunes, and Mattei (Citation2014).
10 For example, in the data, 5 firms received the loan in 2010 and we only have access to their employment for the next year (2011).
11 For more details about CEM method see Iacus, King, and Porro (Citation2011).