Skip to main content

Table 4 Tobit and Heckman regression analysis of the determinants of remittances to the family

From: Why do migrants remit? Testing hypotheses for the case of Morocco

  Two- stage model Tobit
  Selection equation Remittance equation  
  Coef. Robust SE Coef. Robust SE Coef. Robust SE
Use of the fast transfer service 1.034 0.2617***     
Migrant characteristics
Female 0.2035 0.2504 -0.510 0.2641** -0.4320 0.2658
Age 0.1526 0.0545*** 0.0030 0.03933 0.0600 0.0355*
Age squared -0.001 0.0006** 0.000 0.00041 -0.0005 0.0003
Single -0.3166 0.2170 0.1519 0.23849 -0.0495 0.2109
Divorced 0.6778 0.4808 -0.7613 0.3740** -0.5830 0.4063
Widowed     . 0.3986 1.1149
Higher education -0.1487 0.3849 -0.0339 0.33931 -0.0976 0.3236
High secondary education 0.3531 0.3374 0.0210 0.27043 0.11431 0.2830
Middle secondary education 0.1724 0.3264 0.1730 0.26050 0.17320 0.2681
Primary 0.1173 0.2967 -0.3931 0.21874* -0.3907 0.2233*
Duration of stay overseas -0.0136 0.0140 0.0192 0.01203 0.0172 0.0125
Unemployed -2.3766 0.6959*** -0.0120 0.46797 -0.6831 1.1122
Inactive -1.89 0.3339*** -0.3475 0.51967 -1.0911 0.316***
Household characteristics
Expenditure 1.199 0.6305* 0.1869 0.6439 0.5342 0.53263
Expenditure squared -0.2463 0.1327* - 0.0655 0.13089 -0.1391 0.10988
Household head is a male -0.6850 0.2407*** -0.1117 0.2298 -0.3878 0.1907**
Rural 0.1794 0.21762 -0.2488 0.19021 -0.1816 0.18060
Number of dependents living in Morocco -0.1822 0.36561 0.5288 0.3434* .41463 0.32780
Control variables
Number of migrants -0.4802 0.1506*** -0.4281 0.140*** -0.5133 0.134***
Financing of emigration 0.5966 0.1976*** 0.0392 0.2066 0.2247 0.1778
Negative shock 0.3823 0.2858 -0.3167 0.2540 -0.1557 0.2339
Local poverty rate 0.0185 0.0355 -0.000 0.0330 0.00471 0.0299
Constant -3.62 1.439 8.2619 1.447   
Heckman‘s lambda (a)    -0.594 0.3041*   
R-squared 0.396   0.3676    
Number of observations 338    215   
  1. Notes: variable “Widowed” was omitted in the first estimation due to collinearity problem. Coefficients are after correction of heteroscedasticity. ***, ** and *denote significant at thresholds of 1%, 5% and 10% respectively. (a) The p-value is equal to .052. The table reports marginal effects of Tobit estimates.