Understanding Society User Support

Hello,

I am a PhD student at the Centre for Health Economics at University of York.

I am currently using BHPS dataset and I encounter an issue with the variable wFIHHMN: household income, month before the interview.

It seems that this variable has some observations (634) with income zero and others with very low income per month, i.e. less than £10. I have read the BHPS documentation regarding the no imputation of the data on those households where no individual interviews were completed and I thought those zeros might correspond to those cases. However, BHPS usually assigns negative values (i.e. -9, -2, etc.) to those cases instead of zeros.

Therefore, I was wondering whether is there any explanation on why there are entire households with income zero?

Thanks,
Irene.