Capacity to pay calculation

Dear all, I am having some difficulties in the calculation of capacity to pay to estimate catastrophic health expenditure. These were the steps I followed in Stata 16. ****Capacity to pay calculation********* STEP 1: Generate the food expenditure share (share_FE_CTP) for each household by dividing the household’s food expenditure (expfdeqsize_annual_CTP) by its total expenditure (ey_annual). Stata command: gen share_FE_CTP= expfdeqsize_annual_CTP/ey_annual***** STEP 2: Identify the food expenditure shares of total household expenditure that are at the 45th and 55th percentile across the whole sample, name these two variables as food45 and food55. Stata command: _pctile share_FE_CTP, p (45 55) return list *****scalars: scalars: r (r1) = .6567004323005676 r (r2)=.6955645680427551********* STEP 3: Calculate the weighted average of food expenditure (FE) in the 45th to 55th percentile range. This gives the subsistence expenditure per (equivalent) capita, which is also the poverty line (pl). Stata command: tabstat expfdeqsize_annual_CTP if share_FE_CTP > 0.6567004323005676 & share_FE_CTP < .6955645680427551 gen FE = 370.607******** STEP 4: Lastly, calculate the subsistence expenditure for each household (se) as: Stata command: gen SE=FE*eqsize mean FE SE *ctph= exph−foodh if seh > foodh gen Capacity_to_pay= ey_annual-FE****** In this calculation I am observing capacity to pay values below 0, (negative), did I make any any error in the calculation? For more information, please see the following references: Xu, K. Distribution of Health Payments and Catastrophic Expenditures. WHO Geneva. 2005. EIP/HSF/DP.05.2 https://www.who.int/health_financing/documents/dp_e_05_2-distribution_of_health_payments.pdf****** Ghiasvand, Hesam, et al. "Catastrophic health expenditure among Iranian rural and urban households, 2013-2014." Iranian Red Crescent Medical Journal 17.9 (2015).******* With Regards