Ln(Value) = b0Ln(RENT) + b1Ln(Pop) + b2Ln(GDP) + b3Ln(Costs) + b4Ln(Construction

Ln(Value) = b0Ln(RENT) + b1Ln(Pop) + b2Ln(GDP) + b3Ln(Costs) + b4Ln(Constructions) + b5Ln(City_Size) + u(1)whereValue = real house prices
rent = rent values
pop = population
GDP = gross domestic product
Costs = Construction costs
City_Size = the size of the city based on its population.The above model has also been tested with other independent variables like Births Immigrants Net Immigration and Population Change.My question is as follows:
rents population construction costs were find to be statistical significant some at the 1% level some at the 5% level.The above model was regressed twice.
The first time we used rents which are normal average rents while the second time we used rent prices but for houses that were occupied for first time. So we have rents and RentFO The other variables remained the same.The difference from the use of the two different Rent variables is located in the demographic variables and more specific in the variables births which from insignifcant becomes significant to house prices at the 5% level the number of immigrants which from 5% level becomes statistical significant at the 1% level the same goes for Net immigration follows the Immigrants Variables. In addition the population based of the city becomes significant only when we have Rents of first occupancy.So my question is what does this mean and how it is explained? What;s the relationship between the demographic variables and the rentFO that makes the demographic variable to become more significant to house prices?Note construction costs remain significant at the 1% for both models and GDP is insignificant.One last thing is that in both cases just to make it clear rent (normal or FO) are the indepedent variable and that house prices remains the dependent one.