Table 3. Open in new tab Baseline models...

Dependent variable = $Δ \ln Q_{i t}$
Δln P	0.00916*	0.00773	0.00929*	0.00925*
	(0.00481)	(0.00483)	(0.00483)	(0.00480)
Δln CC	−0.0510	−0.0457	−0.0515	−0.0513
	(0.0639)	(0.0645)	(0.0634)	(0.0635)
Obs	840	840	840	840
Instrument validity
First-stage F statistics	133.6	150.3	151.3	78.94
P-value of Over Iden				0.90

Dependent variable = $Δ \ln Q_{i t}$
Δln P	0.00916*	0.00773	0.00929*	0.00925*
	(0.00481)	(0.00483)	(0.00483)	(0.00480)
Δln CC	−0.0510	−0.0457	−0.0515	−0.0513
	(0.0639)	(0.0645)	(0.0634)	(0.0635)
Obs	840	840	840	840
Instrument validity
First-stage F statistics	133.6	150.3	151.3	78.94
P-value of Over Iden				0.90

Table 3.

Baseline models and instruments.

Dependent variable = $Δ \ln Q_{i t}$
	(1)	(2)	(3)	(4)
VARIABLES	Tourist IV	Labor-1 IV	Labor-2 IV	Tourist IV+Labor-2 IV
Δln P	0.00916*	0.00773	0.00929*	0.00925*
	(0.00481)	(0.00483)	(0.00483)	(0.00480)
Δln CC	−0.0510	−0.0457	−0.0515	−0.0513
	(0.0639)	(0.0645)	(0.0634)	(0.0635)
Obs	840	840	840	840
Instrument validity
First-stage F statistics	133.6	150.3	151.3	78.94
P-value of Over Iden				0.90

Dependent variable = $Δ \ln Q_{i t}$
	(1)	(2)	(3)	(4)
VARIABLES	Tourist IV	Labor-1 IV	Labor-2 IV	Tourist IV+Labor-2 IV
Δln P	0.00916*	0.00773	0.00929*	0.00925*
	(0.00481)	(0.00483)	(0.00483)	(0.00480)
Δln CC	−0.0510	−0.0457	−0.0515	−0.0513
	(0.0639)	(0.0645)	(0.0634)	(0.0635)
Obs	840	840	840	840
Instrument validity
First-stage F statistics	133.6	150.3	151.3	78.94
P-value of Over Iden				0.90

Notes: The model is $Δ \ln Q_{i, t} = β_{1} Δ \ln P_{i, t - m} + β_{2} Δ \ln {C C}_{i, t - n} + θ_{t} + ε_{i, t} .$ Our interest is the average supply elasticity estimate, which is $β_{1}$ ⁠. The endogenous $Δ \ln P$ is instrumented by Δln Tourist IV (column 1), Δln Labor-1 IV (column 2), and Δln Labor-2 IV (column 3). The first-stage F statistics reject weak instruments in all columns. The over-identification test is conducted for Tourist IV and Labor-2 IV (column 4), and its high P-value suggests no evidence to reject either instrument, at least from an econometric perspective. Time dummies $θ_{t}$ are not reported. Robust standard errors are in parentheses.

***

P < 0.01.

P < 0.05.

P < 0.1.

This Feature Is Available To Subscribers Only