Table 3.
Panel data regression models of the impacts of low-cost carrier on New Zealand's domestic tourism (June 2009–July 2015).
| Dependent variables |
Model 1 |
Model 2 |
||
|---|---|---|---|---|
| ln(Domestic guest nights)it |
(%) of Domestic guest nightsit |
|||
| Explanatory variables | Model - 1A |
2SLS Model - 1B |
2SLS Model - 1C |
2SLS Model - 2A |
| Coefficients | Coefficients | Coefficients | Coefficients | |
| Constant | -2.0913** (-2.428) |
-2.1247** (-2.415) |
-0.0627 (-1.015) |
0.0032 (0.282) |
| ln(Domestic guest nights)it-1 | – | – | 0.9395*** (82.407) |
– |
| % of (Domestic guest nights)it-1 | – | – | – | 0.9691*** (68.127) |
| ln(LCC's ASK)it | 0.8268*** (16.302) |
0.8305*** a (15.885) |
0.0493*** a (4.128) |
– |
| % of ln(LCC's ASK)it | – | – | – | 0.2320*** b (2.618) |
| ln(GDP per capita)t | 5.9124*** (3.136) |
11.5800*** (3.881) |
0.7709 (0.532) |
0.3545 (0.330) |
| ln(RTI -Accommodation)it | 0.1877*** (5.705) |
0.2155*** (4.742) |
0.3342*** (5.924) |
0.0245 (1.336) |
| ln(RTI -Food & Beverage)it | 0.1753*** (6.455) |
0.2889*** (5.746) |
0.3843*** (6.010) |
0.0405** (2.131) |
| ln(Aviation fuel price)t | 0.0135 (0.119) |
-0.1731 (-0.961) |
-0.0375 (-0.507) |
-0.0317 (-0.765) |
| ln(Petrol price)t | 1.1410*** (3.183) |
0.8084** (2.141) |
0.2831 (1.348) |
0.1503 (1.140) |
| Exchange rate (NZD vs.USD)t | 0.2665 (0.539) |
-0.8195 (-1.046) |
-0.3787 (-1.354) |
-0.0775 (-0.625) |
| Global financial crisis 2008/09 | 0.1872** (2.091) |
0.0754 (0.795) |
0.0074 (0.917) |
-0.0014 (-0.160) |
| Christchurch earthquakes 2011 | -0.0349 (-0.892) |
-0.0796 (-1.773) |
-0.0040 (-0.695) |
0.0011 (0.188) |
| Seasonal dummy(1) | 0.1272*** (4.756) |
0.0743 (1.402) |
0.1168** (2.308) |
0.0144 (1.190) |
| Seasonal dummy(2) | 0.0258 (0.565) |
-0.0101 (-0.160) |
-0.2066*** (-2.853) |
-0.0141 (-1.195) |
| Seasonal dummy(3) | -0.0180 (-0.345) |
0.0046 (0.075) |
-0.0691 (-1.578) |
0.0421*** (3.667) |
| Seasonal dummy(4) | 0.0057 (0.084) |
0.0093 (0.116) |
-0.0413 (-1.383) |
0.0657*** (5.900) |
| Seasonal dummy(5) | -0.0940 (-1.626) |
-0.1945** (-2.298) |
-0.2047*** (-3.786) |
0.0528*** (4.901) |
| Seasonal dummy(6) | -0.1060 (-1.608) |
-0.2179** (-2.259) |
-0.0398 (-0.896) |
0.0415*** (3.742) |
| Seasonal dummy(7) | -0.0451 (-0.588) |
-0.0352 (-0.380) |
0.0455*** (2.584) |
-0.0005 (-0.048) |
| Seasonal dummy(8) | -0.0543 (-0.636) |
-0.0401 (-0.354) |
-0.0928*** (-2.750) |
0.0086 (0.771) |
| Seasonal dummy(9) | -0.0837 (-1.674) |
-0.1220 (-1.788) |
-0.0450 (-1.188) |
-0.0017 (-0.151) |
| Seasonal dummy(10) | -0.0485 (-1.132) |
-0.0714 (-1.066) |
-0.0353 (-0.791) |
0.0056 (0.500) |
| Seasonal dummy(11) | -0.0072 (-0.172) |
-0.0118 (-0.195) |
-0.0696 (-1.593) |
-0.0415*** (-3.754) |
| Adjusted R2 | 0.896 | 0.877 | 0.985 | 0.933 |
| F-statistics | 147.936 | 122.206 | 10.37.99 | 243.440 |
| Observations | 341 | 341 | 341 | 365 |
Remarks: ** and *** indicate that the explanatory variable is significant at the 0.05 and 0.01 significance level, respectively. t-statistics are printed in parentheses. The results of Hausman test verified that all the models favour using the random-effect models.
in Models 1B and 1C represent the predicted values of ln(LCC's ASK)it, which are computed by using the instrumental variable ln(HHI Index)it during the first-stage regression analysis. Results for the first-stage regression analysis are unreported for the sake of brevity.
The results of the Hausman test verified that (%) of ln(LCC's ASK)it in Model 2A is not an endogenous variable.