Dear Amin: after having included fixed effects (FE), a reduction in the explanatories' coefficients makes sense.

I guess you included individual fixed effects. In this case, a significant covariance level between FE and explanatory variables could happen, for several reasons. 

You should base your conclusions on the significant levels (p-value).  Another strategy you could use is to estimate marginal effects. 

Best

Rodolfo

Hi Amin,

This is quite a difficult question to answer without given us more information about the variables you are dealing with. Notwithstanding, i believe the answer to this question is scale dependent and until we know more about how you measured your variables, we may not be able to answer why you are having small coefficients. for example, are all your variables in ratios or are they in logs? Is your model a log-linear model or semi-log?

thanks you Elliot Boateng for sharing your knowledge. actually i am examine the impact of different variables such Non-Performing Loan (npl), Capital Adequacy Ratio (car), Provision for Loan Losses (ppl)on efficiency. The data was collected from annual reports based PKR (Pakistani Rupees) and all the variables are measured in Ratios/Percentage.i am dealing with panel having 300  observations and i used Simple Fixed Effect Model as it is suggestion by Hausman test. i already check different assumption of Fixed effect such normality, autocorrlation, heteroskedasticity etc etc,