Dynamics of EU food safety certification: a survival analysis of firm decisions

Agricultural and Food Economics

Table 4 Results of the means comparison, logit, and survival analyses to explain initial adoption

	Mean	Means comparison		Logit model^a	Survival analysis^b
	All respondents	Certified once	No HACCP	Logit model^a	Survival analysis^b
Incentive
Output price (US$/kg)	3.44	4.12	2.90	0.32* (0.14)	1.44** (0.42)
EU–US price difference (US$/kg)	−0.25	−0.35	0.00	−0.03 (0.43)	0.99 (0.87)
EU–others price difference (US$/kg)	1.29	1.25	1.37	−0.10 (0.22)	0.86 (0.79)
Price of labor (US$/day)	2.93	2.95	2.87	−0.03 (0.23)	1.03 (0.97)
Price of materials (US$/kg)^c	2.14	2.46	1.32*
Price of capital (US$/unit of capital)	0.19	0.21	0.14**	0.06 (0.32)	0.00 (0.00)
Number of markets (number)	3.00	4.00	1.00***	0.02 (0.11)	1.11 (0.87)
Number of product forms (number)	2.00	2.00	3.00*	−0.06 (0.21)	0.74* (0.32)
Output (000 tons/year)	2.60	3.60	0.10**	0.09* (0.03)	1.01** (0.25)
Capacity
Difficulty in credit access (dummy = 0, 1)	0.52	0.30	0.63**	0.04 (0.04)	0.81 (0.96)
Years in business (number)	10.40	11.25	7.00***	0.01 (0.02)	0.96 (1.01)
Active in associations (dummy = 0, 1)	0.58	0.70	0.11**	0.07 (0.05)	2.08**** (0.59)
Number of associations (number)^d	1.40	2.00	0.22**
Age of plant facility (number)	9.30	10.20	8.30	0.02 (0.03)	0.96 (0.87)
N				59	59
Pseudo R squared				0.36
% correctly predicted				60%
Schoenfeld residual					0.32

Source: various firm interviews conducted by the authors. The test of difference in means was performed using a means comparison test (t test) of unequal variances (using both Satterthwaite’s and Welch’s formulas) in the Stata program. Figures in parentheses are the standard errors
***Significant difference in averages at <0.01 level of significance; **at <0.05 level of significance; *at <0.10 level of significance
^aFigures presented here are marginal effects
^bFigures presented here are hazard ratios
^cDue to a multicollinearity problem from the high correlation between output price level and price of materials, we were able to use only one of these variables at a time in the models. However, the two models showed similar results. The one presented here is the one using output price level
^dDue to a multicollinearity problem from the high correlation between “number of association memberships” and “being active in these associations,” we were able to use only one of these variables at a time in the models. However, the two models showed similar results. The one presented here is the one using “being active in these associations”