Dear all,
I'm having a little difficulty creating a model for my analysis. First a little background on my research: I’m researching the relationship between development aid on growth. The hypothesis is that development aid can have a negative influence on a country’s growth due to the so called “dutch disease” effects. Development aid can drive up the exchange rate of a country, causing the export sector of the country to lose competitiveness because the country’s products are now more expensive to other countries. This then results in lower growth.
The model I’m using is given in the following link
(i=industry, j=country). For sake of simplicity I have left out the country and industry fixed effects and control variables. I took the model from a research paper by Rajan & Subramanian (2009)
Basically what the model does (equation 1) is it compares growth between different industries within the country. In the model, the dummy variable “E” is 1 for industries who are export dependent. If development aid results in Dutch disease effects, then industries who are export dependent will grow relatively slower. Thus, alpha is expected to be negative.
Also, the variable “overvaluation” is added to the model (equation 2), to determine whether slower growth of export-dependent industries is explained by exchange rate overvaluation (caused by the Dutch disease). Beta is therefore also expected to be negative, and adding this term to the regression should reduce the coefficient alpha somewhat.
Now comes my question: what I’m trying to research is whether Dutch disease effects can be mitigated by allocating development aid towards the tradable sector. The hypothesis is that in countries where development aid is allocated mostly towards the tradable sector, that export-dependent industries should suffer less loss of growth than in countries where aid is mostly invested within the non-tradable sector. Thus, to test this I would like to add an additional variable (either a dummy or continuous variable) for the “share of aid invested in the tradable sector”.
I’m not sure how to add this to the model. I’ve made an attempt in equation (3), where T is a dummy variable which is 1 for countries who have a share of aid invested in the tradable sector above the median. If my hypothesis is correct, then gamma and delta should be positive, compensating for the negative coefficients alpha and beta (exportable sectors should now suffer relatively less loss of growth).
Hope I could get some feedback on my model, thanks in advance!
I'm having a little difficulty creating a model for my analysis. First a little background on my research: I’m researching the relationship between development aid on growth. The hypothesis is that development aid can have a negative influence on a country’s growth due to the so called “dutch disease” effects. Development aid can drive up the exchange rate of a country, causing the export sector of the country to lose competitiveness because the country’s products are now more expensive to other countries. This then results in lower growth.
The model I’m using is given in the following link
(i=industry, j=country). For sake of simplicity I have left out the country and industry fixed effects and control variables. I took the model from a research paper by Rajan & Subramanian (2009)
Basically what the model does (equation 1) is it compares growth between different industries within the country. In the model, the dummy variable “E” is 1 for industries who are export dependent. If development aid results in Dutch disease effects, then industries who are export dependent will grow relatively slower. Thus, alpha is expected to be negative.
Also, the variable “overvaluation” is added to the model (equation 2), to determine whether slower growth of export-dependent industries is explained by exchange rate overvaluation (caused by the Dutch disease). Beta is therefore also expected to be negative, and adding this term to the regression should reduce the coefficient alpha somewhat.
Now comes my question: what I’m trying to research is whether Dutch disease effects can be mitigated by allocating development aid towards the tradable sector. The hypothesis is that in countries where development aid is allocated mostly towards the tradable sector, that export-dependent industries should suffer less loss of growth than in countries where aid is mostly invested within the non-tradable sector. Thus, to test this I would like to add an additional variable (either a dummy or continuous variable) for the “share of aid invested in the tradable sector”.
I’m not sure how to add this to the model. I’ve made an attempt in equation (3), where T is a dummy variable which is 1 for countries who have a share of aid invested in the tradable sector above the median. If my hypothesis is correct, then gamma and delta should be positive, compensating for the negative coefficients alpha and beta (exportable sectors should now suffer relatively less loss of growth).
Hope I could get some feedback on my model, thanks in advance!
