Table 6. The impact of piece rate and heat exposure (measured in WBGT) on workers’ effort (units in 0.01 METs) based on regression results using a matched sample (Values in parentheses are standard errors).

	Dependent variable: Metabolic rate
	(1)	(2)	(3)
WBGT		4.41	-1.94
		(2.54)	(3.64)
WBGT^2		-0.15	0.00012
		(0.051)	(0.072)
Piece rate	6.98		-142.31
	(3.52)		(49.43)
Piece rate: WBGT			15.64
			(4.41)
Piece rate: WBGT^2			-0.38
			(0.097)
Time of day	0.73	0.78	0.81
	(0.033)	(0.045)	(0.049)
Time of day^2	-0.00038	-0.00039	-0.00040
	(0.000018)	(0.000021)	(0.000023)
Female	-13.97	-12.18	-12.25
	(3.50)	(3.51)	(3.48)
BMI	-0.38	-0.38	-0.35
	(0.38)	(0.37)	(0.38)
Age	-0.21	-0.24	-0.19
	(0.17)	(0.16)	(0.16)
Hired by contractors	0.95	0.0102	-0.51
	(4.08)	(4.28)	(4.27)
Task and Month fixed effects	Yes	Yes	Yes
R-square (without fixed effects) (%)	27.2	28.8	29.2
R-square (with fixed effects) (%)	30.2	31.0	32.2
N. of workers	216	216	216
N. of worker hour	2341	2341	2341

Notes: Other covariates include nonlinear hours of day, task types, age, BMI, hire type, gender and month of the year. The standard error is worker-ID cluster-robust standard error. Column 1 represents results from regression model without no heat exposure variables, while including the pay arrangement variable, and all fixed effects. Column 2 represents results from the regression model without the pay arrangement variable but including heat exposure variables and fixed effects. Column 3 includes both heat exposure variables, pay arrangement variables, their interactions, and all fixed effects.