Effect of the Kyoto Protocol on the demand for natural gas in select OECD countries

Svrha ovog rada je empirijski analizirati potražnju za prirodnim plinom na uzorku od odabranih zemlja OECD-a u periodu od 1995. do 2019. godine te vidjeti jeli međunarodni sporazum o smanjenju klimatskih promjena (protokol iz Kyota) imao utjecaj na njenu potražnju. Analiza je provedena na temelju tri panel modela: model združene regresije, model sa fiksnim efektima i modelom sa slučajnim efektima. Zavisna varijabla u svim modelima je bila potrošnja prirodnog plina, a nezavisne su: BDP, inflacija energije, potražnja za drugim izvorima energije, svjetske rezerve prirodnog plina, direktne strane investicije, udio urbanog stanovništva u ukupnom te binarna varijabla koja prikazuje vrijeme kada je protokola iz Kyota bio na snazi u odabranim zemljama. Odabir najboljeg modela se vršio na temelju tri testa: F-test, Hausmanov test i Breusch-Paganov LM test. Kao najbolji model odabran je model sa fiksnim efektima u kojem su bile značajne varijable BDP, potražnja za drugim oblicima energije i udio urbanog stanovništva. BDP i udio urbanog stanovništva su imali pozitivan utjecaj na potražnju za plinom, dok je potražnja za drugim oblicima energije imala negativan utjecaj. Rezultati provedenog istraživanja su pokazali kako protokol iz Kyota nije imao statistički značajan učinak na potražnju za prirodnim plinom. Iako varijabla nije bila značajna ona je imala negativan utjecaj na potražnju za prirodnim plinom, odnosno njegova potražnja je u prosjeku manja u odnosu na razdoblje kada protokol nije bio zakonom vezujući.The purpose of this paper is to empirically analyze the demand for natural gas in a sample of select OECD countries from 1995. to 2019. and to see whether the international treaty on climate change mitigation (The Kyoto Protocol) had an impact on the demand for natural gas. The analysis was based on three panel models: the pooled regression model, the fixed effects model, and the random effects model. The dependent variable in all models was natural gas consumption, the independent variable are: GDP, energy inflation, demand for other energy sources, worlds natural gas reserves, foreign direct investment, share of urban population in total and a dummy variable showing the time when the Kyoto protocol was in force. The selection of the best model was made on the basis of three tests: F-test, Hausman test and Breusch-Pagan LM test. The best model was the fixed effects model. GDP, demand for other forms of energy and share of urban population were significant in the model. GDP and share of urban population had a positive impact on gas demand, while demand for other forms of energy had a negative impact. The results of the study showed that the Kyoto Protocol did not have a statistically significant effect on the damand for natural gas. Although the variable was not significant, it had a negative impact on the demand for natural gas, ie its demand is on average smaller compared to the period when the protocol was not legally binding