Mixed Logit Model — Reputational harm: all sample vs firm_large = 1, side by side

	(1) Harm — All	(1) Harm — firm_large = 1	(2) Treatment — All	(2) Treatment — firm_large = 1	(3) Treatment + Harm — All	(3) Treatment + Harm — firm_large = 1	(4) Treatment x Harm (Triple) — All	(4) Treatment x Harm (Triple) — firm_large = 1	(5) Subsample (T = 0) — All	(5) Subsample (T = 0) — firm_large = 1	(6) Subsample (T = 1) — All	(6) Subsample (T = 1) — firm_large = 1
* p < 0.1, ** p < 0.05, *** p < 0.01
Cost	−0.348***	−0.600***	−0.216***	−0.240***	−0.374***	−0.620***	−0.530***	0.562***	−0.322***	−0.583***	−0.228**	−0.624***
	(0.067)	(0.104)	(0.040)	(0.058)	(0.066)	(0.105)	(0.075)	(0.113)	(0.091)	(0.137)	(0.094)	(0.163)
Effective	−0.020	0.531**	0.217**	0.314**	−0.020	0.551**	0.126	0.714**	−0.032	0.267	0.145	0.576*
	(0.150)	(0.231)	(0.106)	(0.154)	(0.154)	(0.239)	(0.196)	(0.294)	(0.209)	(0.304)	(0.212)	(0.347)
Socbenefit	0.470***	0.541***	0.467***	0.504***	0.465***	0.541***	0.472***	0.482***	0.450***	0.548***	0.488***	0.481***
	(0.050)	(0.083)	(0.052)	(0.078)	(0.051)	(0.083)	(0.053)	(0.079)	(0.072)	(0.112)	(0.074)	(0.119)
Envbenefit	0.502***	0.532***	0.490***	0.499***	0.503***	0.533***	0.502***	0.475***	0.603***	0.678***	0.366***	0.391***
	(0.046)	(0.071)	(0.045)	(0.066)	(0.045)	(0.071)	(0.046)	(0.068)	(0.066)	(0.100)	(0.062)	(0.105)
Partner_NGO	−0.053	−0.208*	−0.047	−0.179	−0.048	−0.208*	−0.052	−0.152	−0.110	−0.414**	0.008	0.034
	(0.078)	(0.119)	(0.077)	(0.112)	(0.077)	(0.120)	(0.079)	(0.115)	(0.107)	(0.161)	(0.110)	(0.175)
Partner_Gov	−0.018	−0.207	−0.017	−0.153	−0.013	−0.209	−0.019	−0.205	−0.032	−0.416**	−0.003	0.053
	(0.090)	(0.138)	(0.089)	(0.131)	(0.090)	(0.138)	(0.092)	(0.135)	(0.126)	(0.192)	(0.127)	(0.200)
Out	−4.066***	−2.357***	−3.831***	−3.305***	−4.010***	−2.243***	−5.110***	−0.723	−4.644***	−2.904***	−3.633***	−1.385
	(0.636)	(0.718)	(0.470)	(0.590)	(0.621)	(0.753)	(0.742)	(0.986)	(0.958)	(1.102)	(0.898)	(0.948)
first	0.016	0.026	0.022	−0.001	0.017	0.027	0.018	0.011	0.025	−0.016	0.000	0.061
	(0.048)	(0.074)	(0.047)	(0.070)	(0.048)	(0.075)	(0.049)	(0.073)	(0.066)	(0.102)	(0.068)	(0.111)
Cost × Effective	0.106***	0.142***	0.087***	0.119***	0.104***	0.140***	0.102***	0.070*	0.120***	0.187***	0.066*	0.128**
	(0.027)	(0.041)	(0.027)	(0.040)	(0.027)	(0.042)	(0.028)	(0.040)	(0.037)	(0.056)	(0.038)	(0.064)
Cost × negpublicity	0.060**	0.145***			0.057**	0.144***	0.126***	−0.293***	0.031	0.114**	0.041	0.168***
	(0.026)	(0.040)			(0.025)	(0.040)	(0.032)	(0.053)	(0.036)	(0.053)	(0.036)	(0.059)
Effective × negpublicity	0.111**	−0.127*			0.090*	−0.130*	0.034	−0.104	0.062	−0.097	0.129*	−0.098
	(0.051)	(0.077)			(0.050)	(0.077)	(0.073)	(0.109)	(0.072)	(0.103)	(0.070)	(0.113)
Out × negpublicity	0.011	−0.427*			0.051	−0.472*	0.624***	−1.052***	0.287	−0.033	−0.175	−0.795**
	(0.168)	(0.240)			(0.167)	(0.244)	(0.236)	(0.332)	(0.247)	(0.327)	(0.233)	(0.361)
Cost × treatment			0.060	0.023	0.079**	0.044	0.405***	−0.913***
			(0.037)	(0.054)	(0.038)	(0.060)	(0.115)	(0.192)
Effective × treatment			0.131*	−0.032	0.116*	−0.009	−0.111	0.022
			(0.070)	(0.101)	(0.070)	(0.107)	(0.237)	(0.355)
Out × treatment			0.198	0.078	0.158	−0.066	1.809**	−1.002
			(0.238)	(0.345)	(0.242)	(0.361)	(0.788)	(1.163)
Cost × TreatxThreatCon							−0.146***	0.378***
							(0.051)	(0.083)
Effective × TreatxThreatCon							0.103	−0.013
							(0.103)	(0.150)
Out × TreatxThreatCon							−0.834**	0.302
							(0.336)	(0.502)
Num.Obs.	10314	4896	10368	4896	10314	4896	10314	4896	5202	2358	5112	2538
RMSE	0.97	0.97	0.97	0.97	0.97	0.97	0.97	0.97	0.97	0.97	0.97	0.97