Home

full version - Institute for Fiscal Studies

1. The following commands change options for the consumer demand module set_consmod value for consmod set_changewithinc value for inc_aff_spend set_demandparam filename for indata_demand set_demandprice filename for indata_prices set_demandwelfare filename for logfiledemand_welfare 43 set_demandspending filename for logfiledemand_spending set_demandrevenue filename for logfiledemand_revenue Finally the following commands change options for the module which allows incomplete pass through of indirect changes to consumer prices set_indir value for indir set_propprices value for prop_prices set_propwages value for prop_wages set_adout_ind filename for outdata_ad_ind set_logdec_ind filename for logfiledecile_ind set_loghh_ind filename for logfilehhtype_ind set_logrev_ind filename for logfilerevenue_ind Any global not listed above e g NUMGOODSQUAIDS cannot be changed interactively but must be amended directly in main_module do B 2 params_X do The user defines the parameters of the VAT IVA duties IEPS income tax ISR and Social Security tax systems in the parameter do files When naming the do files they should always be of the form params_X do where X is the name of the particular system and corresponds to sys1 sys2 or sys3 as defined in cnterface do Existing parameters modules should be used as templates and it is important that
2. Household containing a single person aged over the pension age Household containing a couple with at least one aged over the pension age Household containing another family grouping e g a multi generational family where there is no child aged under 18 Household containing another family grouping e g a multi generational family with one or more child aged under 18 Household containing at least one member who is not related to the other members e g a family who have live in servants where there is no child aged under 18 Multi family with lt 18 Household containing at least one member who is not related to the other members e g a family who have live in servants with one or more child aged under 18 Users who wish to amend these household types would have to edit the data creation code see Section 6 5 2 The labor module log file The next output to examine is the output of the labor supply module This produces a number of tables that show the impact of the reforms on total formal sector earnings and the impact of this change in earnings on tax revenues given the assumed labor supply elasticities and the average and marginal effective tax rates people face Figure 17 shows part of the logfile logfile_labor in this case labor_practise txt The first column shows pre reform gross income the second the change in formal sector employment income and the final column the change in tax revenu
3. Instructions 1 assign the root program directory 2 assign the input output and parameter directories 3 assign the name for the logfiles 4 assign the name of the input dataset 5 assign the name of the output dataset t t amp Assign default directories global taxsim C LATAX Do_Files global in C LATAX Input_Data global out C LATAX Output_Data global param C LATAX Params Data Assign default log files global logfiledecile decile practise txt Global logfiledecilel decileburden_practise txt global logfilehhtype hhtype practise txt global logfilerevenue revenue _practise txt Assign default input and output files global indata_hh Input_Data_hh global indata_ad Input_Data_ad global indata_co Input _Data_co global outdata_ad Output_adlevel practise global outdata_ hh Output_hhlevel practise 14 The next stage is to change the various options which tell the program what type of analysis you want to perform see Figure 8 The global disttype tells LATAX whether to conduct the distributional analysis using a range of equivalence scales to adjust for household size disttype 1 or only the standard equivalence scale disttype 2 where second and subsequent adults or children aged 12 or over are weighted as 0 8 of the first adult and children aged 11 or under are weighted as 0 5 of an adult For this practise set it equal to 1 The next global netincometype tells LAT
4. 123580 20 scalar BAND6 249243 48 scalar BAND 392841 96 scalar BANDS 1000000000 wkkekkkKEX INCOME TAX RATES scalar RATE1 0 0192 scalar RATE2 0 064 scalar RATES 0 1088 scalar RATE4 0 16 scalar RATES 0 1792 scalar RATE6 0 2136 scalar RATE7 0 2352 scalar RATES 0 3 The changes to the tax system parameters to simulate the originally proposed 2010 system can be made in the same manner the full list of reforms is available in Appendix A But to save time for this practise exercise you can just use the pre defined params_20101 system 4 2 Setting up main_module options The next step is to set the instructions in main_module do to ensure the correct input files are used output files are named and stored correctly and the right program options are used This can be done either by making changes to main_module do or using interactive commands typed into the Stata command line We first describe how to set the options by amending main_module do directly which is particularly useful when changing the default options and then how to set things up using the interactive commands Amending main_module do directly Figure 6 shows the default directory and filename options in LATAX when downloaded from the internet The first thing you might want to do is change the location of the LATAX program and data files To do this you amend the taxsim in out and param globals For instance because LATAX has been
5. D 2 Econometric and empirical specification The QUAIDS is estimated using a 2 step procedure programmed in STATA with standard errors calculated using a clustered bootstrap procedure Because total expenditure may be endogenous we instrument for it using monetary income This is done using a control function approach Stage 1 Before estimation a p and b p are unknown For this reason Ina p is approximated using the Stone price index In p X wi lIn pi i and b p is approximated as 1 Conditional upon the price indices QUAIDS is linear in parameters Hence a linear Seemingly Unrelated Regression SURE framework is used to estimate the model Adding up is imposed by excluding the equation for the nth good from the estimated system of equations parameters for this equation are calculated using the parameters from the other n 1 equations and the adding up restrictions Homogeneity and symmetry are imposed using linear restrictions on parameters Stage 2 The parameters estimated in the first stage are used to calculate values for a p and b p The model is then re estimated using the same specification as the first stage except that p is replaced with a p and by ne The new parameter values used to update a p and b p and the model is then re estimated for a third time This updating of price indices and re estimation is iterated 12 times by which point the parameter values have converged to 5 decimal places Sta
6. Top three marginal rates increase to 21 36 23 52 and 30 in 2010 2011 2012 with a phased reduction to 28 in 2014 Individuals earning up to 6 minimum wages are not affected 2 VAT Impuesto al Valor Agregado IVA General rate of 15 and 10 in border areas General rate of 16 and 11 in border areas 3 Excise duties Impuesto especial sobre la producci n y servicios IEPS 3 a Tobacco 160 rate Additional flat rate of 0 04 for each cigarette or 0 75 Additional flat rate of 0 04 for each grams of snuff to be increased to 0 10 by 2014 cigarette or 0 75 grams of snuff to be increased to 0 10 by 2014 3 b Beer 25 rate 28 rate 26 5 rate temporary 3 c Lottery 20 rate 30 rate 30 rate 3 d Drinks with alcohol 50 rate Additional minimum charge per litre of 3 pesos 53 rate content greater than 20 by volume 3 e Telecommunications None 4 rate 3 rate except for Internet connexions 4 New expenditure tax Introduction of a 2 expenditure tax on all goods and Rejected Contribucion para el Combate a la Pobreza services with the exception of the purchase of government licenses and donations to charity 5 Tax on cash deposits 2 rate of balance 3 rate of balance 3 rate of balance Source CEFP 2009f 39 Appendix B This appendix provides details on the main_module do module and the parameters X do parameters files so t
7. global logfiledecile ind decile ind p75 _w75_practise txt global logfilehhtype ind hhtype ind p75 w75_practise txt global logfilerevenue_ind revenue ind p75 w75 practise txt control parameter don t change global indic incidence 1 If this option is turned on you must tell LATAX the proportion of any indirect tax changes that are passed through to prices prop_prices and the proportion of the indirect tax change that is borne by firms i e not passed through to prices that is ultimately borne by workers in the form of lower wages as opposed to being borne by shareholders prop_wages For this practise exercise let us assume that both proportions are 0 75 In order words 75 of any changes in indirect tax will be passed through to prices of the 25 borne by firms 75 will be borne by workers in the form of lower wages The next globals define the adult level output file and the log files for the analysis with less than full pass through of indirect tax outdata_ad_ind logfiledecile_ind logfilehhtype_ind 17 and logfilerevenenue_ind It is a good convention to ensure that the names include an _ind element to make sure you don t overwrite the files created for the analysis assuming full pass through For this practise lets also add _p75_w75_practise to the end of each file name to indicate that 75 of indirect tax changes are passed into prices 75 of the tax borne by firms is passed into wages and that th
8. 1 Food 2 83 0 Non alcoholic beverages 3 68 1 Alcohol and tobacco 11 87 2 Clothing and footwear 1 34 3 Housing electricity gas water and other 7 65 fuels 4 Furnishings and household equipment 2 55 5 Health 4 69 6 Transport 3 55 7 Leisure and Culture 5 24 8 Education 1 08 9 Hotels cafes and restaurants 1 91 10 Miscellaneous goods and services 3 46 Table C 5 Adult Individual Output File Variables Variable Name Description folioviv Residence identifier foliohog Household identifier numren Individual identifier sex Sex age Age education Education level empstat Employment status incl to inc20 Gross income by source as in table C2 hhtype Household type indicator numfam Number of families in household residentes Number of residents in household mayores Number of adult household members menores Number of household members under 12 years old eqsc_5030 Equivalence scale eqsc_8050 Equivalence scale eqsc_100100 Equivalence scale factor Sample Weight estrato Strata Identifier ubica_geo Geographic Identifier 57 totexp Annual total expenditure monetary and non monetary in Mexican monexp Annual monetary expenditure in Mexican nmonexp Annual non monetary expenditure in Mexican moninc_e Annual net total current income monetary non monetary in Mexican moninc Annual net monetary current income in Mexican totinc Annual net
9. The last column shows the average total gain or loss for households in each decile group Scrolling down the next two tables show the impact of the second set of reforms system 20101 on the same basis i e of total expenditure total expenditure decile groups and a per capita equivalence scale The impact of the two sets of reforms is then shown as a proportion of monetary expenditure mexp total income tinc and monetary income minc in turn again using total expenditure decile groups and a per capita equivalence scale Next outputs ate produced showing gains losses as a proportion of total expenditure texp monetary expenditure mexp total income tinc and monetary income minc in turn this time using monetary expenditure decile groups and per capita equivalence scales This is then repeated for total income decile groups and monetary income decile groups Lastly everything is repeated for two other equivalence scales First an equivalence scale where the first adult is given a weight of 1 subsequent adults and children aged 12 or over are given a weight of 0 8 and children aged 11 or under are given a weight of 0 5 es85 Then an equivalence scale where the first adult is given a weight of 1 subsequent adults and children aged 12 or over are given a weight of 0 5 and children aged 11 or under are given a weight of 0 3 es53 This latter scale is that used in the UK and is termed the modified OECD scale Becaus
10. VAT exempted VAT exempted VAT zero rate VAT zero rate VAT zero rate VAT zero rate Tobacco Health services Education services Lottery Public school transport and transport abroad Transfers and other services Leisure goods and services Household good and services Food and drinks Education goods Household good and services Leisure services Gastodiario Gastos Gastoeduca Gastos Gastodiario Gastos Gastos Gastos Gastodiario Gastos Gastos Gastos Clave A239 A240 amp lug_com gt 3 Clave J001 J004 J005 J007 J008 J013 J016 J017 J018 J036 J039 J062 J072 amp lug_com lt 4 Clave E001 E008 E015 E017 T905 amp lug_com lt 4 Clave E029 amp lug_com lt 4 Clave B001 B007 E013 M001 T902 amp lug_com lt 4 Clave G002 G006 G011 N006 N007 N011 N016 amp lug_com lt 4 Clave E020 E022 E026 amp lug_com lt 4 Clave G012 G019 amp lug_com lt 4 Clave A001 A068 A070 A072 A197 A203 A215 A218 A242 amp lug_com lt 4 Clave E014 amp lug_com lt 4 Clave G007 amp lug com lt 4 Clave L029 amp lug_com lt 4 54 expnum3 expnum9 excluded expnum11 expnum8 none excluded expnum11 expnum6 expnum1 expnum9 expnum6 expnum11 Taxed Alcohol and tobacco Non taxed health and education Leisure and hotels Non taxed transport goods and services and petrol Leisure and
11. full pass through of indirect tax Creates output files and recalls dirtax and taxbase Quaids Module Applies demand model calculates expenditure patterns and welfare effects Expenditure Pattern File Recalls Indirtax household and distribution modules Labor Module Uses elasticities Labor Supply File PTRs and METRs to calc revenue effect of changes in labor supply main_module do This interface main module contains user edited instructions on the directories in which the simulation code input and output data and parameter code can be found the name and location of logfiles the names of input and output datasets the existing and reform systems to be used in the analysis sys1 sys2 sys3 and runs the tax simulator There are options for whether one wants to conduct the behavioral analysis A worked example explaining how to use the main_module can be found in section 4 whilst full information on all the options can be found in Appendix B loadprogs do This module calls the programs that perform the tax revenue and distributional calculations and that estimate the behavioral response to a particular set of reforms given the assumptions provided to LATAX This is done as follows e First for each of the three tax systems chosen in main_module do sys1 sys2 sys3 load the parameters file params_X from the correct parameters folder and then call calc_indirtax do set_taxbase do and calc_dirtax do e Then
12. our input data does not identify which workers work for firms or which individuals are owners shareholders of companies who would be affected by a particular set of indirect tax changes Instead we change all income from wages and salaries by a fixed proportion so that the total change in wages and salaries is equal to the fraction of the indirect tax change borne by workers and all income from profits or dividends by another fixed proportion so that the total change in income from these sources is equal to the fraction of the indirect tax change borne by shareholders Unless the relevant workers or shareholders are distributed significantly differently to workers and shareholders more generally this is likely to give at least a rough idea how less than full pass through would affect the distributional and revenue impact of reforms Second the columns for the gains losses due to income tax and social security contributions are replaced with a single column showing the change in net income This is because less than full pass through of indirect tax changes implies gross income is affected by the reforms This means net income is affected by not only changes in tax payments but also changes in gross income and it is therefore clearest to show the impact of these two changes together rather than separately The log files produced are logfiledecile_ind in this case decile_ind_p75_w75_practise txt logfilehhtype_ind in this case hhtype_ind_
13. security through work trabajos 1 formal IMMS 2 formal ISSSTE 3 formal PEMEX inc1 formal Gross Wages and salaries ingresos clave P001 annual pobla08 P002 P003 income in trabajos P006 P011 Mexican P015 2008 inc2 formal Gross Overtime ingresos clave P004 annual pobla08 income in trabajos Mexican 2008 inc3 formal Gross End of the year bonus ingresos clave P009 annual pobla08 48 income in trabajos Mexican 2008 inc4 formal Gross Incentives rewards and prizes ingresos clave P005 annual pobla08 income in trabajos Mexican 2008 inc5 formal Gross Holiday bonuses and cash ingresos clave P007 annual allowances pobla08 income in trabajos Mexican 2008 inc6 formal Gross Profit sharing ingresos clave P008 annual pobla08 P019 income in trabajos Mexican 2008 inc7 formal Gross Pensions ingresos clave P032 annual pobla08 income in trabajos Mexican 2008 Inc8 formal Gross Other labor income ingresos Clave P013 annual pobla08 P017 P018 net income in trabajos of P019 Mexican 2008 inc9 formal Net annual Capital ingresos Clave P016 income in pobla08 P023 P031 Mexican trabajos P047 P065 2008 inc10 formal Net annual Other income ingresos inc_semp income in pobla08 clave P067 Mexican trabajos P080 inc_tran 2008 clave P033 P045 P066 inc_other clave P046 inc_oemp clave P020 P022 P063 P064 inc11 informal Gross Wages and salaries ingreso
14. 0 069 0 015 0 041 0 010 0 009 2 Food on which IVA is levied and meals out 0 077 0 847 0 072 0 010 0 255 0 041 0 068 0 122 0 043 0 137 0 011 0 011 3 Alcoholic Drinks and Tobacco IVA and IEPS levied 4762 1 390 0 744 0 370 0 410 0 210 0 343 0 325 0 205 0 172 0 416 0 225 4 Clothing and footwear IVA levied 0 181 0 043 0 044 1 049 0 530 0 070 0 125 0 026 0 018 0 063 0 052 0 014 5 Household goods services and communications IVA levied IEPS sometimes levied 6 Household goods services and communications no IVA levied 0 035 0 254 0 073 0 207 0 031 1 050 0 065 0 312 0 004 0 043 0 168 0 049 7 Transport and vehicle fuels IVA levied IEPS sometimes levied but not modelled 8 Public Transport and other transport on which no IVA levied 0 364 0 286 0 041 0 037 0 223 0 108 0 041 1 217 0 050 0 112 0 036 0 009 9 Health and Education goods no IVA levied 0 199 0 191 0 018 0 183 0 886 0 006 0 027 0 083 0 029 0 108 0 059 0 020 0 180 0 246 0 058 0 166 0 184 0 041 0 841 0 051 0 069 0 048 0 154 0 004 0 204 0 189 0 050 0 030 0 166 0 001 0 079 0 106 0 954 0 012 0 085 0 034 10 Health and personal goods and services IVA levied 0177 0 276 0 018 0 058 0 269 0 014 0 031 0 094 0 005 0 934 0 019 0 001 11 Leisure and hotel services IVA sometimes levied 9497 0 047 0 108 0 087 0 428 0140 0 206 0 075 0 102 0 046 1 038 0 057 12 Other services 0 697 0 360 0 374 0 214 0 821 0 231 0 044 0 133 0 238 0
15. 005 0 366 0 932 Notes Standard errors have not yet been calculated due to the time the bootstrapping process takes An updated version of this table will be provided when this has been completed Elasticities are estimated using mean prices and expenditures and for a household with 2 adults and 2 children where the head is male has low levels of education is employed and lives in the DF Source Authors calculations using MEXTAX Bank of Mexico price data and ENIGH 2008 68 Table D 4 Income elasticities Good Income Elasticity 1 Food on which no IVA is levied 0 52 2 Food on which IVA is levied and meals out 1 34 3 Alcoholic Drinks and Tobacco IVA and IEPS levied 1 16 4 Clothing and footwear IVA levied 1 20 5 Household goods services and communications IVA levied IEPS sometimes levied 1 20 6 Household goods services and communications no IVA levied 0 84 7 Transport and vehicle fuels IVA levied IEPS sometimes levied but not modelled 2 06 8 Public Transport and other transport on which no IVA levied 0 66 9 Health and Education goods no IVA levied 1 12 10 Health and personal goods and services IVA levied 0 98 11 Leisure and hotel services IVA sometimes levied 2 09 12 Other services 1 69 Notes Standard errors have not yet been calculated due to the time the bootstrapping process takes An updated version of this table will be provided when this has been completed Elasticities are estimated using
16. 387 124 632 Decile 8 135 926 209 o 478 154 842 Decile 9 166 351 438 o 637 197 1 272 Richest l 324 373 1 340 o 1 339 324 3 003 ae ee ee eae et eae eee eee oe a ee eee ee eee Average 118 121 224 o 408 115 747 IMPACT OF REFORMS AS A PROPORTION OF texp USING DECILES OF texp AND EQUIVALENCE SCALE es11 FOR REFORM SYSTEM 20100 Decile texp ISR Soc Sec IVA IEPS TOTAL EE ee eee ote a a eee ec ea a E aa Poorest 38 626 0 00 0 00 0 22 0 04 0 26 Decile 2 57 940 0 01 0 00 0 25 0 05 0 32 Decile 3 70 275 0 01 0 00 0 27 0 07 0 35 Decile 4 78 243 0 02 0 00 0 28 0 08 0 38 Decile 5 90 825 0 04 0 00 0 30 0 09 0 43 Decile 6 102 458 0 07 0 00 0 31 0 10 0 48 Decile 7 116 323 0 10 0 00 0 33 0 11 0 54 Decile 8 135 926 0 15 0 00 0 35 0 11 0 62 Decile 9 166 351 0 26 0 00 0 38 0 12 0 76 Richest 324 373 0 41 0 00 0 41 0 10 0 93 Danae iar PPE RER N aaa ae ea aNenetnemnannenkanaanauaca owas asus Average 118 121 0 11 0 00 0 31 0 09 0 51 The first column shows the wneguivalised average expenditure or income for each decile group the next four columns show the average mean gain or loss for households in each decile group 23 from changes in the amount of income tax social security VAT IVA and duties IEPS paid
17. 4 Clave J002 J003 J006 J011 J012 J015 J019 J040 J041 J043 J060 J065 J071 amp lug_com lt 4 Clave F010 F011 F012 amp lug_com lt 4 Clave F001 F002 F003 F004 F005 F006 F008 F009 amp lug_com lt 4 Clave A228 A231 A234 A232 A238 amp lug_com lt 4 Clave A226 A237 amp lug_com lt 4 Clave A223 A225 A227 A229 A230 A233 A235 A236 amp lug_com lt 4 Clave A224 amp lug_com lt 4 Clave A239 A240 amp lug_com lt 4 expnum12 expnum10 expnum7 expnum5 expnum3 expnum3 expnum3 expnum3 expnum3 Other taxed and non taxed such as financial services Taxed personal goods and services including taxed health education etc Taxed transport goods and services and petrol Taxed household goods services communications and electronics Taxed Alcohol and tobacco Taxed Alcohol and tobacco Taxed Alcohol and tobacco Taxed Alcohol and tobacco Taxed Alcohol and tobacco Note This category is assumed to be VAT zero rate for the purpose of integrating the consumer demand system in the simulator See explanations to table C 1 below This means that the estimates of IVA revenues are slightly smaller but the difference is negligible since these drink categories are very small in terms of total expenditure 56 Table C 4 National Account Consumption Categories and Factors Category Number Description Multiplication Factor
18. H136 T908 amp lug_com gt 3 Clave N001 N002 T914 amp lug_com gt 3 Clave J002 J003 J006 J011 J012 J015 J019 J040 J041 J043 J060 J065 J071 amp lug_com gt 3 Clave F010 F011 F012 amp lug_com gt 3 Clave F001 F002 F003 F004 F005 F006 F008 F009 amp lug_com gt 3 Clave A228 A231 A234 A232 A238 amp lug_com gt 3 Clave A226 A237 amp lug_com gt 3 Clave A223 A225 A227 A229 A230 A233 A235 A236 amp lug com gt 3 Clave A224 amp lug_com gt 3 53 none excluded expnum11 expnum7 expnum4 expnum12 expnum10 expnum7 expnum5 expnum3 expnum3 expnum3 expnum3 Leisure and hotels Taxed transport goods and services and petrol Taxed clothing and footware Other taxed and non taxed such as financial services Taxed personal goods and services including taxed health education etc Taxed transport goods and services and petrol Taxed household goods services communications and electronics Taxed Alcohol and tobacco Taxed Alcohol and tobacco Taxed Alcohol and tobacco Taxed Alcohol and tobacco exp30 exp31 exp32 exp33 exp34 exp35 exp36 exp37 exp38 exp39 exp40 exp41 Formal Informal Informal Informal Informal Informal Informal Informal Informal Informal Informal Informal VAT and duties VAT exempted VAT exempted VAT exempted VAT exempted VAT exempted
19. Income 5 3 The consumer demand module log files The consumer demand module produces a number of log files The first is the distributional file ogfiledemand_welfare which shows the impact of the reforms to indirect taxes only by decile group and by household type changes due to direct taxes have to be added on Figure 18 shows the first part of the output of this file Separate tables show the gains losses in cash and proportional terms and using different variables construct decile groups and different equivalence scales The second column shows the cash or proportional gains losses when one accounts for the ability of consumers to substitute between goods This is measured by the compensating variation of the reform The third column shows the cash or proportion gains losses assuming no behavioural response In this instance allowing for demand effects makes little impact on the size of the losses to households Note that neither column is directly comparable to the figures contained in the main logfiledecile outputs because some approximation is required when conducting consumer demand analysis If one wants the most accurate no behavioural response figures then the results of logfiledecile should be used rather than those from the logfiledemand_welfare 30 Figure 18 The distributional impact of reforms including consumer demand responses IMPACT OF REFORMS IN CASH TERMS USING DECILES OF totexp USING EQUIVALENCE SCALE eqscll FOR
20. REFORM SYSTEM 2010 Decile totexp with Behavior without Behavior Poorest 38 626 103 103 Decile 2 57 940 182 183 Decile 3 70 275 245 246 Decile 4 78 243 288 289 Decile 5 90 825 364 365 Decile 6 102 458 431 432 Decile 7 116 323 514 515 Decile 8 135 926 620 622 Decile 9 166 351 811 813 Richest 324 373 1 676 1 677 SSE SR S AAS Beene Ce AE E E ee re eee Average l 118 121 524 524 IMPACT OF REFORMS IN PROPORTIONAL TERMS USING DECILES OF totexp USING EQUIVALENCE SCALE eqscil FOR REFORM SYSTEM 2010 Decile totexp with Behavior without Behavior SAE eS ee Be eet ee ee er ets Poorest 38 626 0 27 0 27 Decile 2 57 940 0 31 0 32 Decile 3 70 275 0 35 0 35 Decile 4 78 243 0 37 0 37 Decile 5 90 825 0 40 0 40 Decile 6 102 458 0 42 0 42 Decile 7 116 323 0 44 0 44 Decile 8 135 926 0 46 0 46 Decile 9 166 351 0 49 0 49 Richest l 324 373 0 52 0 52 R Average 118 121 0 40 0 40 The next logfile produced is logfiledemand_spending which shows the impact of the reforms on spending patterns Note that in this example we have set the global inc_aff_spend 1 which means that in the change in income resulting from the change in direct taxes as well as the change in prices resulting from changes in indirect taxation is allowed to impact upon spendi
21. T901 A223 A241 clave gt HO001 amp clave lt H136 clave T908 65 Bank of Mexico PRICES codes 1 100 103 110 101 102 111 115 378 381 116 122 134 171 Notes Prices for A215 sill bottled water zero rate and A216 sparkling water iva is levied on this are together so include in category 1 as A215 represents a bigger expenditure share in ENIGH 2008 A218 bottled juices or cordials zero rate and A217 prepared water and natural juices iva is levied on this are together include in as A218 represents a bigger expenditure share in ENIGH 2008 No specific price for A242 food dispensed by government agencies or NGOs No prices for jewlery wrist watches and other women accessories Banxico code 172 ENIGH code H125 H127 H129 H131 include them and use price of 171 handbags Household goods services and communications IVA levied IEPS sometimes levied Household goods services and communications no IVA levied Transport and vehicle fuels IVA levied IEPS sometimes levied but not modelled Public Transport and other transport on which no IVA levied Health and Education goods no IVA levied Health and personal goods and services IVA levied expnums expnum6 expnum7 expnums expnum9 expnum10 clave gt C001 amp clave lt C024 G009 G010 G020 G022 1001 1026 K001 K044 T903 T907 T909 T911 electronics L001 L022 F001 F009 G008 G007 G01
22. able to be luxuries i e having an income elasticity of greater than 1 at one level of total expenditure and necessities i e having an income elasticity of less than 1 at another Banks Blundell and Lewbel 1997 The model assumes that the utility obtained from any particular good is not affected by the amount one works and therefore demand for goods is also unaffected and it does not allow for positive or negative externalities from expenditure on certain goods for instance fuel alcohol and tobacco The Quadratic Almost Ideal Demand System QUAIDS is a generalisation of the Almost Ideal Demand System AIDS and is based on the following indirect utility function TE Ina p t 1 a xo Where x is expenditure a p b p and A p are defined as 1 Ina p a ailn P 5 gt yy np In pj i i j b p ija i i 1 Ind p X ailn pi i 1 where i 1 n denotes a good Applying Roy s identity this gives the following equation for wi the share of expenditure on good i in total expenditure is for each household 7 2 neat Innes le For the resulting demands to be consistent with utility maximisation the demand system must satisfy four key properties adding up homogeneity symmetry and negativity negative semi definiteness The first three can be imposed using linear restrictions on the parameters of the model adding up 60 n n n a 1 gt Bi 0 X y OVI gt 4 0 i l i
23. at stake here is whether a household s position in the income distribution or expenditure distribution gives a better indication of whether they are rich or poor In order to assess this one must first understand that household surveys generally pick up a snapshot measure of income or expenditure e g income in the last month or spending on different types of items in periods ranging from one week to one year But such a short term measure might not accurately reflect the living standards of the household in either the short or long run For instance households with low incomes may be able to use borrowings savings or previously purchased durable goods to maintain their living standards at least in the short run Many economists have argued that households should be ranked by their consumption as this takes account of such smoothing of income shocks Porterba 1989 Meyer and Sullivan 2003 2004 2008 2011 The argument for using consumption is particularly persuasive if we believe households smooth their consumption over long periods of time and we are concerned with the long term distributional impact of a policy change This might suggest a preference for using expenditure to rank households But expenditure is not the same as consumption expenditure captures the purchase costs of durable goods like cars whereas consumption captures the flow of benefits from these goods Like income expenditure may be volatile with househo
24. described above The first thing to do is to set the directories for the input output and parameters data You do this by typing the following commands into the Stata command prompt and pressing enter after each line set_inputdir C LATAX Input_Data set_outputdir C LATAX Output_Data set_paramdir C LATAX Params_Data This overwrites the default options only until the defaults are re set so is handy if you want to make a temporary change to an output folder Given that we are using the default input files there is no need to change these But we do want to change the output file names Input and execute the following commands set_adout output_adlevel_practise set_hhout output_hhlevel_practise and the name of the log files for distributional and revenue analysis set_logdec decile_practise txt set_logbur decileburden_practise txt 18 set_loghh hhtype_practise txt set_logrev revenue_practise txt Figure 10 shows how the bottom of the main Stata window and command prompt line should look when you are entering the last of these commands Figure 10 Directories and filenames inputted interactively using Stata commands AO SA set_systems 2008 20100 20101 defaults CLA Input Data Ci LAT utput_ Data C LATAX Params_Data output_adlewel_practise set_hhout output_hhlewel_practise decile_practise decileburden_practise Command set_logrev revenue
25. each income source and broad expenditure category to match National Accounts see the LATAX website for these To adjust gross income to match national accounts when creating your own dataset towards the end of the create_grossincome do file lines should be added multiplying the relevant income sources by the relevant factor As with the analysis produced in Abramovsky et al 2011 one may decide to have factors that vary across the income distribution or that are related to the extent to which expenditure is under recorded or vice versa To adjust household net income to match national accounts when creating your own dataset amend a block of code should be inserted just before create different categories gen double moninc_e inc_emp 12 36 in create_LATAX_input_data do that multiplies each individual income source by the relevant factor For example replace inc_emp inc_emp 1 6173 1 1 replace inc_semp inc_semp 2 5191 1 1 replace inc_cap inc_cap 26 0441 1 1 replace inc_tran inc_tran 1 2948 replace inc_oemp inc_oemp 1 6173 1 1 replace inc_other inc_other is the code inserted in the do file that creates the alternative input dataset with adjustment for under recording that we have made available for download from the LATAX website To adjust monetary expenditure for under recording in ENIGH code should be inserted in create_processed_data do just prior to when each monetary expenditure file is saved separate
26. from the ENIGH horages dta file If the user wishes to rename these variables or create a single identifier they can do so and then amend the global identlist in main_module do of LATAX For instance if a single variable called ident were created that uniquely identified each household this would be changed to global identlist ident The variable numren identifies individuals and in the household file is set equal to 1 allowing the household and adult level files to be merged when required The hhtype variable is a household type indicator created in create_processed_data do which is used to produce summary distributional analysis of reforms Three equivalence scales are required eqsc_5030 eqsc_8050 and eqsc_100100 and these cannot be renamed without changing the main LATAX code It is not advised to change what scale each variable refers to e g eqsc_5030 should be used to store the value of the scale where a first adult is given a weight of 1 second and subsequent people aged 12 or over are given a weight of 0 5 and children aged 11 or under a eight of 0 3 but one can do so if one wishes These are calculated in create_LATAX_input_dta do The variable factor is the grossing factor used to weight the sample so that it matches the national Mexican population It is taken directly from the hogares dta ENIGH file The variable ubica_geo is a geographical identifier which is taken from the hogares dta ENIGH file and is us
27. in to prices These tell the program whether to run the module with less than full pass through set_indir 1 means yes the proportion of the tax change passed through to prices set_propprices and the proportion of that not passed through that impacts on wages set_propwages as well as the output files and log files Figure 12 shows these commands once they have been entered Once this is done all necessary commands have been input and the model can be run by typing simulator into the Stata command line and pressing enter Section 5 explains how to interpret the results of the simulator once they have been outputted Before that Section 4 3 provides information for readers who are interested in changing the built in elasticities of labor supply used in the labor supply model built into LATAX other readers can skip to Section 5 20 Figure 12 Setting less than full pass through indirect tax options interactively using Stata commands set_indir 1 set_propprices 0 75 set_propwages 0 75 set_adout_ind out_adlevel_ind_p75_w75_practise dta set_logdec_ind decile_ind_p75_w75_practise txt set_loghh_ind hhtype_ind_p75_w75_practise txt set_logrev_ind revenue_ind_p75_w75_practise txt Command simulator 4 3 Changing labour supply elasticities The labor supply model built in to LATAX allows for elasticities of employment and hours of work earnings to vaty by sub group of the population To prevent the assum
28. is predicted to occur from the relevant reform to the tax system References Abramovsky L O Attanasio C Emmerson and D Phillips 2011 The distributional impact of reforms to direct and indirect tax in Mexico Analytical Report and Results Institute for Fiscal Studies IFS in consortium 2011 A retrospective evaluation of the EU VAT system Report prepared for the EU Commission Meyer B and J Sullivan 2003 Measuring the Well Being of the Poor Using Income and Consumption Journal of Human Resources 38 S1180 1220 Meyer B and J Sullivan 2004 The Effects of Welfare and Tax Reform The Material Well Being of Single Mothers in the 1980s and 1990s Journal of Public Economics 88 July 1387 1420 Meyer B and J Sullivan 2008 Changes in the Consumption Income and Well Being of Single Mother Headed Families American Economic Review 98 5 December 2221 2241 Meyer B and J Sullivan 2011 Further Results on Measuring the Well Being of the Poor Using Income and Consumption Canadian Journal of Economics 44 1 52 87 Poterba J 1989 Lifetime Incidence and the distributional burden of excise taxes American Economic Review 79 pp325 30 71
29. mean prices and expenditures and for a household with 2 adults and 2 children where the head is male has low levels of education is employed and lives in the DF Source Authors calculations using MEXTAX Bank of Mexico price data and ENIGH 2008 69 The patterns of substitution and complementarity seem reasonable Food on which IVA is not levied 1 is a substitute for food on which IVA is levied and meals out 2 2 is a substitute for both 1 and alcohol and tobacco 3 Private 7 and public 8 transport are also substitutes Clothing 4 is complementary with 2 and 3 possibly reflecting additional demand for clothing when one is visiting restaurants and bars and other venues where food and alcohol are served The income elasticities are also sensible Food on which IVA is not levied 1 is a necessity whilst food on which IVA is levied and meals out 2 is a luxury The other strong necessity is public transport whilst private transport leisure goods and services and other services are strong luxuries D 5 Using LATAX with a different consumer demand model It is beyond the scope of this manual to provide instructions on estimating a new consumer demand model However there may be circumstances when one wishes to do so For instance the general demand model used here groups all food commodities together but if one wished to model the application of VA VAT on some but not all food items one might wish to use a model which h
30. sae ERREN SE NEE inn ee eae aoe ne eee eee Poorest 33 795 0 00 0 00 0 33 0 05 0 38 Decile 2 56 039 0 00 0 00 0 28 0 06 0 34 Decile 3 71 422 0 00 0 00 0 29 0 07 0 36 Decile 4 87 847 0 00 0 00 0 26 0 07 0 33 Decile 5 106 375 0 00 0 00 0 26 0 08 0 35 Decile 6 124 121 0 02 0 00 0 27 0 08 0 37 Decile 7 140 826 0 03 0 00 0 27 0 08 0 39 Decile 8 174 744 0 10 0 00 0 28 0 09 0 47 Decile 9 226 157 0 17 0 00 0 29 0 09 0 56 Richest 475 085 0 34 0 00 0 26 0 06 0 66 ses ait ee cae RARE TAANE a an es ca cs esa ae eae se ph ieee cae Average 149 634 0 07 0 00 0 28 0 08 0 42 Comparing the outputs in Figures 14 and 15 one can see how the distributional impact of reforms can look significantly different when households are ranked into decile groups differently and especially when gains losses are expressed as a proportion of income rather than expenditure In this instance the reforms system 20100 look to be significantly more progressive i e the rich bear more of the burden when using total expenditure decile groups and dividing losses by total expenditure Figure 14 than when using total income decile groups and dividing losses by total income Figure 15 In particular the changes to IVA VAT appear to be progressive when expressed as a fraction of expenditure over the e
31. tax lability that occurs when one enters a new tax band This allows one to model entry fees to the tax system such as the minimum 57 14 payment under the 2010 El Salvador system 45 NUMCREDS Number of bands of employment income subsidy LCRED 0 NUMCREDS Upper limit of each employment income subsidy band The 0 band is to ensure that no subsidy is given to those with no earned income ACRED 1 NUMCREDS Amount of credit for those with an income in each subsidy band For social security the user needs to define the following scalars NUMSOCSECSYSTEMS The number of systems with different rates and or bands in operation NUMSOCSEC 1 NUMSOCSECSYSTEMS The number of bands for each of the social security systems For each of the social security systems X 1 NUMSOCSECSYSTEMS and for each of the bands under that system Y 1 NUMSOCSEC 1 NUMSOCSECSYSTEMS the following parameter are also defined LSOCSECX_Y The upper limit of band Y for social security system X SOCSECX_Y The contribution rate applying to band Y for social security system X SSNUMSOURCES Number of sources of income for social security purposes SSEXEMPT 1 SSNUMSOURCES The cash amount of a particular source that is exempt from social security For sources of income on which tax is not being simulated e g capital income this should be set to an amount higher than the largest observed value for that source SSPEXEMPT 1 SSNUMSOURCE
32. using interactive Stata commands The following lists the optional commands and shows how to use them For instance typing set_progdir C LATAX Do_Files changes the taxsim global see above to C LATAX Do_Files The rest of the commands are wtitten more generically in the following style set_progdir directory for taxsim Remember DO NOT type the The following commands can be typed to change the remaining file directories set_inputdir directory for in set_outputdir directory for out set_paramdir directory for param 42 Input and output filenames can be changed using the following commands set_consdat filename for indata_co set_addat filename for indata_ad set_hhdat filename for indata_hh set_adout filename for outdata_ad set_hhout filename for outdata_hh The following commands change the names of the outputted logfiles from the no behavioral response simulations set_logdec filename for logfiledecile set_logbur filename for logfiledecile1 set_loghh filename for logfilehhtype set_logrev filename for logfilerevenue The following commands change options relating to how to carry out analysis set_disttype value for disttype set_netincometype value for netincometype The following commands change options for the labor module set_labmod value for labmod set_labout filename for logfilelabor
33. 1 G019 F010 F017 M002 M004 M006 M012 M018 T906 T913 B001 B007 M001 E013 T902 Health J001 J004 JOOS J007 J008 J013 J016 J017 J018 J036 J039 J062 J072 J009 J010 J014 J020 J035 J037 J038 J042 J044 J059 J063 J064 T910 Education E014 Taxed education E016 E018 E019 Taxed health J002 J003 J006 J011 J012 J015 J019 J040 J041 J043 J060 J061 JO65 JO71 Personal goods and services D001 D026 T904 66 184 186 189 190 195 205 246 294 355 361 185 187 310 318 320 323 325 305 309 256 265 268 269 271 272 276 277 342 343 266 267 270 273 275 278 293 295 344 346 No information on the price level of electricity Banxico code 188 ENIGH code G008 use the price of gas Banxico code 189 No information on the price level of landline telephone services Banxico code 190 and ENIGH code F003 or national inter city phone calls and international calls Banxico codes 191 amp 192 and ENIGH code F002 use the price of internet conexion Banxico code 355 No price information on mobile telephone services ENIGH codes F004 and F005 public telephones ENIGH codes F006 and mail and other communications services ENIGH codes F007 F009 Exclude housing renting cost and imputed rent from this category Banxico codes 182 and 183 and ENIGH codes G001 G006 No price level information for tolls Banxico code and ENIGH code M005 or car insur
34. 2 M005 NO08 N010 T915 amp lug com lt 4 Clave E023 E025 E027 E028 E030 E033 T912 LO23 L028 NO03 N005 amp lug_com lt 4 Clave F013 F017 T906 T913 M002 M004 M006 M018 amp lug_com lt 4 Clave H001 H136 T908 amp lug_com lt 4 55 expnum9 expnum10 expnum1 expnum2 expnum5 expnum10 none excluded expnum11 expnum7 expnum4 Non taxed health and education Taxed personal goods and services including taxed health education etc Non taxed food and drinks Taxed food drinks and food out Taxed household goods services communications and electronics Taxed personal goods and services including taxed health education etc Leisure and hotels Taxed transport goods and services and petrol Taxed clothing and footware exp52 exp53 exp54 exp55 exp56 exp57 exp58 exp59 exp60 Informal Informal Informal Informal Informal Informal Informal Informal Informal VAT taxed VAT taxed VAT taxed VAT taxed VAT and duties VAT and duties VAT and duties VAT and duties VAT and duties Other non food goods and services Health related goods and services Petrol Telecoms under 14 alcohol 14 20 alcohol over 20 alcohol beer tobacco Gastos Gastos Gastos Gastos gastodiario Gastodiario Gastodiario Gastodiario Gastodiario Clave N001 N002 T914 amp lug com lt
35. 4 69 if exptype 5 replace gas_mon gas_mon 4 69 if exptype 5 replace gas_tri gas_tri 3 55 if exptype replace gas_mon gas_mon 3 55 if exptype 6 replace gas_tri gas_tri 5 24 if exptype replace gas_mon gas_mon 5 24 if exptype replace gas_tri gas_tri 1 08 if exptype replace gas_mon gas_mon 1 08 if exptype 8 replace gas_tri gas_tri 1 91 if exptype replace gas_mon gas_mon 1 91 if exptype 9 replace gas_tri gas_tri 3 46 if exptype 10 replace gas_mon gas_mon 3 46 if exptype 10 A list of the National Accounts categories can be found in table C4 in Appendix C 7 Appendices These appendices provide details on the 2010 Mexican reforms being used as an example in this manual They also provide details on input variables output variables and the calculation modules 38 Appendix A Table A 1 A description of the 2010 tax reforms Item Status quo 2010 tax reform proposed by the Executive power 2010 tax reform approved by the Congress and implemented 1 Income tax both personal and corporate Impuesto sobre la Renta ISR Top three marginal rates are 19 94 21 95 and 28 Top three marginal rates increase to 21 36 23 52 and 30 in 2010 2011 2012 with a phased reduction to 28 in 2014 Individuals earning up to 4 minimum wages are not affected The annual upper threshold of income band 3 lower threshold of income band 4 decreases from 88 793 04 mex to 79 964 16
36. 8 Dec 2012 20 44 43 Revenue raised from the tax reforms Tax Reform system 20100 Reform system 20101 ee eee eee ISR 5 987 545 600 8 471 142 912 Soc Sec a 0 IVA 10 888 586 240 38 020 403 200 IEPS 3 064 007 424 4 083 248 640 Total 19 940 139 008 50 574 794 752 name lt unnamed gt log C LATAX output_Data revenue_practise txt log type text closed on 18 Dec 2012 20 44 43 The log reports the change in revenue from each tax in Mexican Pesos following the two sets of reforms analysed A positive value means an increase in revenue and a negative value means a decrease in revenue following the reforms The analysis of the impact of the reforms across the income and expenditure distributions is recorded in the log file declared in logfiledecile in this case decile_practise Gains to 22 households are shown as positive values and losses are shown as negative values This log file includes multiple tables that show the distributional impact of the reform in slightly different ways e Using expenditure or income as the variable by which to rank households into decile groups from poorest to richest e Using expenditure or income to divide cash gains losses into proportional gains losses e Using measures of total income and expenditure that include non monetary amounts such as payment in kind and imputed rent and using measures of monetary income and expenditure
37. AX whether it needs to calculate household net income netincometype 1 or whether it is provided in the household level input dataset indata_hh netincometype 2 If you are using either Mexican input data downloaded from the LATAX website or created input data from ENIGH using the do files provided see Section 6 leave the this global at its default value 2 If you have created your own input dataset from data that does not include net income measures change to 1 identlist lists the variables used to identify households The default is folioviv foliohog for Mexico Only change this if you have created your own input dataset with different household identifiers PTO 7 Also set to 1 if using LATAX with data for El Salvador to model reforms in that country Change this to hhid if using the model for analysis in El Salvador using El Salvadoran data 15 Figure 8 Amended control variables labor supply and consumer demand options CONTROL VARIABLES 1 full range of distributional analyses 2 reduced range global disttype 1 1 use calculated net income 2 use net income from input Global netincometype 2 Set the Household Identifier List Global identlist folioviv foliohog MODULES INCLUDING BEHAVIORAL RESPONSE MARGINS 1 LABOR SUPPLY MODEL do you want to run labor model 1 Yes 2 No DEFAULT SET TO 2 NO global labmod i Others global logfile
38. DUCTSOCSEC 1 The amount of subsidy for employment income is calculated using the tax base for income tax and the ACRED and LCRED amounts defined for the system Once income tax and social security payments are calculated net income is calculated as gross income income tax social security contributions Depending on the value of the global indic_incidence the output is saved either as a standard file indic_incidence 1 or as output data for the analysis of less than full pass through indic_incidence 2 or 3 do_household do This module first loads the relevant variables from the input data individual and household files and then merges in the results of the indirect and direct tax calculations for the base system sys1 and the two reform systems sys2 and sys3 Individual and household level changes in tax payments under the reform systems are then calculated Household net monetary and non monetary income is calculated from the tax simulation outputs if netincometype 1 in the interface main module Individual level and household level output files are saved do_distanalysis do This program first calculates the revenue effects of each reform by summing and appropriately grossing using sample weights the changes in tax payments for each tax This is saved in a log file The program then calculates equivalised incomes for each equivalence scale 100 100 100 100 80 50 and 100 50 30 and equivalised income and expend
39. Formal Informal Formal Formal Formal Formal Formal Formal Tax classification 2008 VAT exempted VAT exempted VAT exempted VAT exempted VAT exempted VAT exempted ENIGH 2008 file Definition Residence identifier Household identifier Health services Gastos Education Gastoeduca services Lottery Gastos Public Gastodiario school transport and transport abroad Transfers Gastos and other services Leisure Gastos goods and services Variable description Clave J001 J004 J005 J007 J008 J013 J016 J017 J018 J036 J039 J062 J072 amp lug_com gt 3 Clave E001 E008 E015 E017 T905 amp lug_com gt 3 Clave E029 amp lug_com gt 3 Clave B001 B007 E013 M001 T902 amp lug_com gt 3 Clave G002 G006 G011 N006 N007 N011 N016 amp lug_com gt 3 Clave E020 E022 E026 amp lug com gt 3 51 Variable name expnum9 excluded expnum11 expnum8 none excluded expnum11 Demand system categories Description Non taxed health and education Leisure and hotels Non taxed transport goods and services and petrol Leisure and hotels exp7 exp8 exp9 exp10 exp11 exp12 exp13 exp14 exp15 exp16 exp17 Formal Formal Formal Formal Formal Formal Formal Formal Formal Formal Formal VAT exempted VAT zero rate VAT zero rate VAT zero rate VAT zero rate VAT zero rat
40. LATAX A multi country flexible tax micro simulation model Institute for Fiscal Studies Laura Abramovsky and David Phillips September 2015 LATAX is a multi country flexible tax micro simulation model developed by researchers from the Institute for Fiscal Studies IFS for the analysis of VAT excise duties income tax and social security contributions as well as non means tested price subsidies It is designed to be a simple public tool for micro analyses of future or counterfactual tax reforms in Mexico El Salvador and potentially other countries LATAX can quantify the revenue and distributional impact of tax reforms under both the assumption that individuals do not change their behavior as a consequence of changes in taxes and the assumption that individuals react to these changes along specific margins LATAX is a user friendly simulator program designed so that users do not need to edit the main program code but can instead make changes in i tax system parameters and ii estimate accounting for behavioral impacts on households income expenditure Based on the available data and the user defined tax parameters the program produces indirect tax payments the direct tax base and direct tax payments before calculating the revenue effects of the reforms and the impact of the tax changes across the income expenditure distributions and by household types Full details of the simulator program can be found in the background papers by Abramovsk
41. S The proportion of a particular source that is exempt from social security SSMPEXEMPT 1 SSNUMSOURCES The amount of a source above which no additional proportion is exempt from social security 46 Appendix C Table C 1 Household file Variables Definition ENIGH 2008 file Variable description folioviv Residence identifier foliohog Household identifier numren Individual identifier This is just to be able to merge results from the adult file collapsed at the household level There is only one observation for each household and numren 01 for all households residents Number of household Hogares members menores Number of household Concen members under 12 years old mayores Number of adult residentes menores household members eqsc_5030 Equivalence scale 1 mayores 1 0 5 menores 0 3 eqsc_8050 Equivalence scale 1 mayores 1 0 8 menores 0 5 eqsc_100100 Equivalence scale 1 mayores 1 1 menores 1 factor Sampling weights Hogares estrato Stratum Hogares ubica_geo Area Identifier Hogares hhtype Household type Pobla08 See hhtype dta above numfam Number of families in the Pobla08 See hhtype dta above household totexp Annual total gastos gastodiarios Sum of all expnum expenditure monetary gastoeduca categories defined in and non monetary in nomonatario hogares consumption file below Mexican 2008 non monetary expenditure irent autocon reminkind tr
42. STATA software which has been designed by IFS researchers to be a public tool for analyses of future tax reforms LATAX can quantify the revenue and distributional impact of tax reforms it calculates taxes paid for each household and produces summary distributional and revenue analysis under both the assumption that individuals do not change their behavior as a consequence of changes in taxes and the assumption that individuals react to these changes along specific margins i Labour supply if suitable elasticities of participation and hours of work are available ii VAT pass through the degree to which indirect taxes are passed on to consumers by producers if suitable estimates of VAT pass through are available iii Consumer demand consumers responses to changes in prices induced by changes in taxes using estimated price elasticities in the case of Mexico LATAX examines the distributional and revenue impacts of tax reforms at one point in time It accurately models rules and structures of complex tax systems It covers i Income tax ii Value Added Taxes iii Excise duties iv Social security contributions LATAX models the tax system at the level of the individual or household using survey or micro level administrative data it requires the use of detailed household level data with detailed information on demographics income and expenditure the precise requirements are discussed in Section 6 Sample weights al
43. This scalar defines the second VAT rate applicable to a particular category in the tax system under consideration This should be changed when changing the parameters of a reform or baseline system DUTIES 1 NUMGOODS _ The duties rate applicable to a particular category in the tax system under consideration This should be changed when changing the parameters of a reform or baseline system For income taxes the user needs to define the following scalars NUMSOURCES The number of income sources in the input data DEDUCTSOCSEC Set equal to 1 if social security contributions are deducted from taxable income as they are in El Salvador EXEMPT 1 NUMSOURCES The cash amount of a particular source that is exempt from income tax For sources of income on which tax is not being simulated e g capital income this should be set to an amount higher than the largest observed value for that source PEXEMPT 1 NUMSOURCES The proportion of a particular source that is exempt from income tax MPEXEMPT 1 NUMSOURCES The amount of a source above which no additional proportion is exempt from income tax NUMBANDS The number of income tax bands BAND 0 NUMBANDS The upper limit of each income tax band The upper limit of the higher band should be higher than the largest observed value for gross income RATE 1 NUMBANDS Income tax rates 0 1 MAXRATE Maximum average tax rate if one applies JUMP 1 NUMBANDS The discrete jump in
44. ULATOR The first matrix is for public sector workers subject to ISSSTE and the second matrix is for private sector works subject to IMSS Checking that gross income is being calculated correctly is likely to be one of the trickiest parts of creating a new input dataset for LATAX and should not be rushed It is worthwhile working out a number of examples whereby one calculates for a given gross income the net income implied by the tax system and then using the data creation code to recover the gross income This should reduce the likelihood of errors 35 6 3 The expenditure files A full list of variables can be found in table C3 of Appendix C All required variables are in bold The expenditure variables are optional in the sense that the number and ordering of items can be changed provided that the parameters and main_module files are correctly amended although it is recommended to maintain the defaults for ease of use This file contains information on monetary expenditures only separately by whether an item is deemed to be purchased formally or informally The default is for the first 30 items to be formal sector items and the next 30 items to be informal sector items Expenditure by a household on an item is classified as informal if the household reports purchasing it from a market stall or street vendor ENIGH variable lug_com 1 2 or 3 LATAX requires annual expenditures meaning that the input data from ENIGH which reco
45. _practise Next the options that tell the program which type of analysis to conduct should be set First we set which type of distributional analysis to conduct and tell the program to use net income from the input files rather than calculating it set_disttype 1 set_netinctype 2 Before turning on labour supply modelling and setting the location of the output file set_labmod 1 set_labout labor_practise txt And then turning on consumer demand modelling and setting the location of the input and output files required for this modelling set_consmod 1 set_changewithinc 1 set_demandparam demand_params dta 19 set_demandprice demand_prices dta set_demandwelfare decile_demand_practise txt set_demandspending spending _demand_practise txt set_demandrevenue revenue_demand_practise txt Figure 11 shows the inputting of the options for these options Figure 11 Setting labor supply and consumption options using Stata commands set_disttype 1 set_netinctype 1 set_labmod 1 set_labout labor_practise txt set_consmod 1 set_changewithinc 1 set_demandparam demand_params dta set_demandprice demand_prices dta set_demandwelfare decile_demand_practise txt set_demandspending spending_demand_practise txt Command set_demandrevenue revenue_demand_practise txt The last set of options to program up are those relating to less than full pass through of indirect tax changes
46. ance Banxico code 319 and ENIGH code N008 exclude them Exclude purchases of automobiles including cars bycicles etc Banxico codes 311 and 312 ENIGH codes M007 M011 No price level information for education fees Banxico codes 335 342 ENIGH codes E001 E008 E017 E015 T905 exclude them from the system Prices for J060 cotton wool dressings iva levied on this and J061 surgical alcohol zero rated are together so include in category 10 as J060 represents a bigger expenditure share in ENIGH 2008 Exclude education services E009 E012 expnum11 Non taxed leisure E020 E022 E026 347 349 Leisure and hotel services IVA sometimes levied Other services E029 L029 Taxed leisure E023 354 356 E025 E027 E028 E030 E033 L023 360 362 L028 N003 N005 T912 364 No specific price for lottery E029 or hotels expnum12 N001 N002 T914 382 384 Table D 2 Other variables used in the demand system Variable name Demographic variables child adults sex empstat educlow educmid central north_interior north_coastal west east south south_east Description Number of household members under 12 years old Number of household members 12 years or over if the head of the household is female 0 otherwise 1 if the head of the household is employed 0 otherwise 1 if the head of the household has primary education or less 0 otherwise if the head of the household has secondary ed
47. and the gain loss to households differs labor do This program allows one to estimate how taxable employment income responds to changes in both direct and indirect taxes given an assumed set of elasticities on the intensive and extensive margin First the household level average tax rates on expenditure are calculated accounting for informal transactions These are then added to average income tax and social security rates to get the participation tax rates PTRs and to marginal income tax and social security rates to get marginal effective tax rates METRs The user then defines the hours elasticities intensive margin and the participation elasticities extensive margin which may vary by demographic group The changes in taxable employment income and the associated changes in tax revenues are then calculated using the calculated PTRs and METRs and the assumed elasticities Finally demographic variables are created and the changes in taxable labor income and revenue are outputted in logfilelabor quaids do This module allows one to estimate how consumer spending patterns change following tax changes and how such changes in spending patterns affect the revenues from the tax changes and consumer welfare It makes use of demand system estimates calculated in quaids_estimation do see Appendix D First the expenditure on each of the 12 demand system goods categories is calculated using the NUMGOODS categories of the MEXTAX ind
48. anges in tax payments net incomes by household income and expenditure decile groups logfiledecile1 is the logfile used to record the proportion of the gains losses following a reform accruing to each household income and expenditure decile group logfilehhtype is the logfile used to record changes in tax payments net incomes by household type as defined in the input data logfilerevenue is the logfile used to record the changes in tax revenues following a reform In this example as this is a practise let us suppose you just add _practise to the end of every filename but before the txt 13 Third the globals input_hh input_ad and input_co define the names of the input files that contain basic household demographics adult demographics and incomes and expenditure information respectively These should be amended if you have named the input files differently or if you want to use a different version of an input file for instance one that corrects for under recording of income or expenditure differently than the standard file In this example let s leave the filenames unchanged Fourth the globals outdata_ad and outdata_hh define the names of the output files that store adult and household level measures of the impact of the reforms Again let us rename these by adding _practise at the end Figure 7 shows what you now should have Figure 7 Amended directory and filename options MAIN PROGRAM program set defaults
49. ansfer defined in nomonet_agg dta above monexp Annual monetary gastos gastodiarios Sum of all expnum expenditure in Mexican gastoeduca categories defined in 2008 consumption file below nmonexp Annual non monetary nomonatario hogares irent autocon expenditure in Mexican reminkind transfer 2008 defined in irent dta and nomonet_agg dta above 47 totinc Annual net total ingresos nomonetario current income hogares monetary non monetary in Mexican inc_emp inc_semp inc_cap inc_tran inc_oemp inc_other irent autocon 2008 reminkind transfer defined in ingresos_mon_net_agg dta and nomonet_agg dta above moninc Annual net monetary ingresos inc_emp inc_semp current income in inc_cap inc_tran Mexican 2008 inc_oemp inc_other see definition in ingresos_mon_net_agg dta moninc_e Annual net employment Ingresos inc_emp see definition in monetary current ingresos_mon_net_agg dta income in Mexican 2008 Table C 2 Adult file Variables Formal Description Definition ENIGH 2008 Variable Informal file description folioviv Residence identifier foliohog Household identifier numren Individual identifier pobla08 age Age pobla08 edad sex Sex pobla08 sexo education Highest qualification achieved pobla08 n_instr161 empstat Whether in work pobla08 Equals 1 if trabajo 1 or verifica 1 2 3 4 5 formal_w Whether covered by social pobla08 0 informal
50. ant thing to note is that ENIGH provides data not on gross income as is required by a tax simulator like LATAX but instead net income This means the data creation programs must calculate gross income from the reported net income To do this we backwards calculate gross income given net income and the tax rates thresholds and exemptions that were in place during the year in which the survey was collected Hence the code contained in create_grossincome do creates gross income from the 2008 ENIGH based on the 2008 tax and social security contributions system If one wishes to use more up to date information one will need to replace the following information in create_grossincome do e The amount of each income source that is subject to taxes and the amount that is exempt with that from the relevant year see code starting after heading Income including exemptions which usually depends upon some multiple of the monthly or daily minimum wage Note that these amount are monthly amounts e The minimum wages with those from the relevant year see code starting after the heading NOW MAKING SURE ALL FORMAL PAID MIN WAGE Note that this should be expressed as an annual amount e The tax rates fixed tax amounts i e amount of tax paid according to tax rates below the current tax rate tax credits tax thresholds and social security contribution rates with those from the relevant year see code starting BACK TO THE GROSS INCOME CALC
51. ar VATS scalar VAT6 scalar VAT7 scalar VATS scalar VATS scalar VAT10 scalar VAT11 scalar VAT12 scalar VAT1i3 scalar VAT14 scalar VAT15 scalar VAT16 scalar VAT1i7 scalar VATi8 scalar VAT19 scalar VAT20 scalar VAT21 scalar VAT22 scalar VAT23 scalar VAT24 scalar VAT25 scalar VAT26 scalar VAT27 scalar VAT28 scalar VAT29 scalar VAT30 Il ooooco ooo oO an p a a p oa a a H j nO oO towed Ha pd p MOA a ll Seeoeeecaceceoecqcoodrooeseaoe amp oO e oO oe M o Figure 4 Duties rates following the 2010 tax reforms scalar DUTIES25 0 03 scalar DUTIES26 0 265 scalar DUTIES27 0 30 scalar DUTIES28 0 53 scalar DUTIES29 0 265 scalar DUTIES30 1 64 The 2010 tax reforms also involved changes to income tax rates The tax rates are only one of the income tax parameters that can be easily varied in LATAX users can also easily change the thresholds exemptions the number of rates and tax credit amounts information on which is available in appendix B To change the tax rates RATE6 RATE7 and RATE8 as shown in Figure 5 This is the final change to model for the 2010 tax system so save the params_20100 do file 11 Figure 5 Income tax parameters following 2010 tax reforms wkkkkKKK INCOME TAX THRESHOLDS scalar NUMBANDS 8 scalar BANDO 0 scalar BAND1 5952 84 scalar BAND2 50524 92 scalar BANDS 88793 04 scalar BAND4 103218 00 scalar BANDS
52. as disaggregated categories of food In principle it is possible to do this but a number of changes would have to be made to LATAX in order to do this First one would have to change the relevant parts of main_module do e The global indata_demand would have to be updated to the name of the input file containing the parameters of the new demand system and the global indata_prices would have to be updated to the name of the input file containing the baseline prices for the goods in the new demand system e Ifthe number of categories included in the demand system is going to be less than or greater than 12 NUMGOODSQUAIDS must be updated accordingly e Ifthe new demand system requires disaggregating the main expenditure categories used in the no behavioral response simulations the data creation programs must be amended accordingly and a new version of indata_co created This will necessitate updating the parameters X do files e The main expenditure categories used in the demand system need to be inputted in NUMGOODSDEM and the expenditure categories in each demand system category need to be defined in the goodslists The final goodslist should list the expenditure categories not included in any demand system category e The categslists should be updated so that each main expenditure category is associated with a demand system category Second if one plans on amending the demographic variables included in the demand system or if one wishe
53. be considered more informative and will be the focus of discussion in the remainder of this chapter although results will be presented as a proportion of income in Annex I for the purpose of completeness The argument that showing VAT payments as a fraction of income may give a misleading impression of the lifetime distributional impact of VAT is driven by the potential for households to borrow and save but it does not rely on households being able to borrow freely or have large amounts of savings to draw down Neither does it rely on consumers being rational and forward looking or engaging in optimal consumption smoothing To see this consider a poor household with a long run income of 100 euros per week but who is currently spending 200 euros per week funded by drawing down the last of their savings Furthermore suppose that the rate of VAT is 25 on all goods and services The household would pay 40 euros per week in VAT equal to 20 of their current spending but 40 of their current income The question is which measure is a better reflection of the impact of VAT on the household It is true that their current income is a better measure of their long run purchasing power than their current expenditure is But it does not follow that expressing VAT payments as a proportion of current income gives a better measure of the impact of VAT on that long run purchasing power This is because when the household is forced to cut their spending back to th
54. call do_household do and do_distanalysis do e Finally if selected using consmod indir and labmod as set in main_module do call quaids do indirect_incidence do and labor do params_X do The user defines the parameters of the VAT duties income tax ISR and Social Security tax systems in the parameter do files When naming the do files they should always be of the form params_X do where X is the numerical name of the particular system and corresponds to sys1 sys2 or sys3 as defined in main_module do Existing parameters modules should be used as templates and it is important that the names of scalars are not changed although of course the values can be A worked example explaining how to amend the params_X modules can be found in section 4 whilst full information on all the parameters can be found in Appendix B Note that the descriptions of the calculation modules below refer to particular parameters of the tax systems defined in Appendix B calc_indirtax do This program first uses the VAT and duties rates that applied at the time of the survey underlying the expenditure input data vectors OVAT OBVAT and ODUTIES to calculate pre tax prices Total pre tax expenditure is also calculated Next the VAT and IEPS rates that apply under the system under consideration vectors VAT BVAT and DUTIES are used to calculate the amount of VAT and IEPS due under that system The calculation is done by good and then summed over goods to give a
55. designed to be a multiple country model the existing default input output and parameters directories include space for a country sub directory e g global in C LATAX country Input_Data If you are planning to use the model only for Mexico and have created the folders as set out in Section 3 simply delete country from each of the in out and param globals 12 Figure 6 Default directory and filename options MAIN PROGRAM program set_defaults Instructions 1 assign the root program directory 2 assign the input output and parameter directories 3 assign the name for the logfiles 4 assign the name of the input dataset 5 assign the name of the output dataset t t Assign default directories global taxsim C LATAX Do Files global in C LATAX country Input_Data global out C LATAX country Output_Data global param C LATAX country Params Data Assign default log files Global logfiledecile decile txt Global logfiledecilel decileburden txt Global logfilehhtype hhtype txt Global logfilerevenue revenue txt Assign default input and output files global indata_hh Input _Data_hh global indata_ad Input _Data_ad global indata_co Input _Data_co global outdata_ad Output_adlevel global outdata hh Output_hhlevel Second you might also want to change the name of the various Stata logfiles created by LATAX logfiledecile is the logfile used to record ch
56. e VAT zero rate VAT zero rate VAT taxed VAT taxed VAT taxed Household good and services Food and drinks Education goods Household good and services Leisure services Health related goods and services Health related goods and services Food and drinks Food and drinks Household goods and services Personal goods and services including education and health Gastos Gastodiario Gastos Gastos Gastos Gastos Gastos Gastodiario Gastodiario Gastos Gastos Clave G012 G019 amp lug_com gt 3 Clave A001 A068 A070 A072 A197 A203 A215 A218 A242 amp lug_com gt 3 Clave E014 amp lug_com gt 3 Clave G007 amp lug_com gt 3 Clave L029 amp lug_com gt 3 Clave J009 J010 J014 JO20 J035 J037 J038 J042 J044 J059 J063 J064 T910 amp lug_com gt 3 Clave J061 amp lug_com gt 3 Clave A069 A071 A216 A217 amp lug_com gt 3 Clave A198 A202 A219 A222 A243 A247 T901 amp lug_com gt 3 Clave C001 C024 T903 GO08 G010 G020 G022 T909 F007 T911 1001 1026 K001 K044 L001 L022 T907 amp lug_com gt 3 Clave D001 D026 T904 E016 E018 E019 amp lug_com gt 3 52 expnum6 expnum1 expnum9 expnum6 expnum11 expnum9 expnum10 expnum1 expnum2 expnum5 expnum10 Non taxed household goods services and electronics Non taxed food and drinks Non taxed health and education Non taxed
57. e a larger proportion of expenditure in Mexico is likely to be on food and other goods that cannot be easily be shared the modified OECD scale may result in too small adjustments to the income for larger households when ranking them into decile groups The es85 scale allows for some but fewer economies of scale whilst the es11 scale implies no economies of scale and that children cost the same as adults Figure 15 shows the impact of the reform system 20100 as a percentage of total income across the total income distribution Figure 15 Dividing gains losses by income or expenditure Decile tinc ISR Soc sec IVA IEPS TOTAL Spee es ae a a ee EE en ees ee el ee ee oe ae ea ee Poorest 33 795 o o 111 18 129 Decile 2 56 039 0 o 159 33 193 Decile 3 71 422 0 o 204 52 256 Decile 4 87 847 1 o 226 63 290 Decile 5 106 375 5 o 280 84 369 Decile 6 124 121 26 o 332 103 461 Decile 7 140 826 49 o 374 5 119 542 Decile 8 174 744 174 5 o 483 164 S 821 Decile 9 226 357 391 o 666 205 1 262 Richest 475 085 1 597 o 1 245 307 3 149 ER E PEE STEA AE E TANS ENE E oe eee ae A Average l 149 634 224 o 408 115 747 IMPACT OF REFORMS AS A PROPORTION OF tinc USING DECILES OF tinc AND EQUIVALENCE SCALE es11 FOR REFORM SYSTEM 20100 Decile tinc ISR Soc Sec IVA IEPS TOTAL ES
58. e level of their long run income 100 euros per week the amount of VAT they would pay falls to 20 euros per week This is equal to 20 of their current and long run income and their long run expenditure of 100 euros per week Measuring VAT payments as a percentage of current spending thus gives a more meaningful measure of VAT s distributional impact than measuring payments as a percentage of current income even in cases where current income is the better measure of long run living standards This demonstrates a key point the best measure of lifetime living standards might be current income for some households and current spending for other households This is a separate issue from the arguments in favour of expressing VAT as a percentage of expenditure instead of income and expressing direct taxes like Income Tax as a percentage of income rather than expenditure a The assumption that lifetime income and expenditure are equal means that we abstract from gifts and bequests This is for ease of exposition only the argument with bequests is more complicated but conclusions are unchanged For example when assessing the proportional impact of VAT on households that are recipients of gifts and bequests it seems clear that we would 27 want to take into account those gifts and bequests when measuring their lifetime resources We would not for instance wish to say that a household with zero income but large expenditures funded by gift
59. ed in consumer demand simulation and estimation The variables totexp monexp nmonexp moninc_e moninc and totinc are calculated variables which record the annual expenditure and income of a household as a whole These are calculated from expenditure on individual items and individual income sources for each person in a household It is possible to have moninc and totinc calculated directly by LATAX rather than having them as part of the household input file To do this the global netincometype in main _module do should be set to 1 Section 6 4 describes how to amend the income and expenditure data so as to roughly account for the under recording of income and expenditure in the ENIGH survey 6 2 The adult input file This file contains information on the demographics and gross incomes of individuals including children as well as adults although only the latter are used in practise A full list of variables can be found in table C2 of Appendix C with all those in bold required and the income variables in italics optional in the sense that the number of income sources can be amended provided that the relevant parameters code is the amended in LATAX We now describe the main required vatiables As with the household file household by default folioviv and foliohog and individual numren identifiers are required and are taken directly from the ENIGH data The program also requires the variables sex age education and rel_head relatio
60. employment monetary current income in Mexican socsec_ sys1 sys2 sys3 Annual social security contributions for each system socrate_ sys1 sys2 sys3 Marginal social security contribution rate for each system inctax_ sys1 sys2 sys3 Annual income tax net of tax credits for each system taxrate_ sys1 sys2 sys3 Marginal income tax rate for each system ieps_tot sys1 sys2 sys3 Household total IEPS paid for each system iva_tot sys1 sys2 sys3 Household total IVA paid for each system hh_change_inctax sys2 sys3 Change in income tax paid between base and each reform system hh_change_socsec sys2 sys3 Change in social security contributions paid between base and each reform system hh_change_ieps sys2 sys3 Change in IEPS paid between base and each reform system hh_change_iva sys2 sys3 Change in IVA paid between base and each reform system hh_change_tot sys2 sys3 Change in total tax paid between base and each reform system hhnet sys1 sys2 sys3 Household net income under each system Table C 6 Household Output File Variables Variable Name Description folioviv Residence identifier foliohog Household identifier numren Individual identifier hhtype Household type indicator eqsc_5030 Equivalence scale eqsc_8050 Equivalence scale eqsc_100100 Equivalence scale factor Sample Weight totexp Annual total expendi
61. es associated with the change in formal sector income Results are shown by age group education group sex and for the population as a whole In order to get the total revenue effects of the reform including the labor supply response one must add the Change in Tax shown in this log file to the change in revenues shown in the logfile_revenue logfile the pre reform revenue change 29 Figure 17 Showing how the burden is borne across the distribution File Edit Format View Help Labor supply effects of reforms by educ group for system 20100 Educ Group l Gross Income Change in Income Change in Tax ee ew ww ww ow we we ow we wo oo wo oo oe ee nee None or Preschool 87 266 590 720 66 302 380 5 250 465 Primary School 634 931 249 152 541 788 352 61 872 012 secondary school 643 106 406 400 638 275 840 102 205 576 Degree inc PhD 670 410 866 688 1 440 441 984 380 247 520 commercial Professional 1 381 925 257 216 3 432 065 536 940 344 960 a ee ee ee ee ee ee ee ee ee a a l E Labor supply effects of reforms by age group for system 20100 sex l Gross Income Change in Income change in Tax enn 4 eee Male 2 326 901 293 056 4 276 269 824 1 062 716 736 Female l 1 090 739 175 424 1 842 604 288 427 203 776 Rn D nnn nr rrr eee Labor supply effects of reforms by age group for system 20100 Change in
62. eutral as it is imposed on all goods and services at the same rate it has the same proportional effect on the purchasing power of rich and poor households VAT payments under such a system would be the same fraction of both lifetime income and lifetime expenditure for rich and poor households 26 But suppose as in reality we only have information on current income and spending If VAT payments are presented as a fraction of current expenditure this distributionally neutral pattern of payments would be found However because households with low current income tend to spend more than their income and those with high current income tend to spend less showing payments as a fraction of net income will make the uniform VAT look regressive if households are defined as rich or poor based on their current income On the other hand if households are defined as rich or poor based on their current expenditure because households with the lowest spending tend to report incomes that are higher than their spending and those with high spending tend to report incomes that are lower than their spending showing VAT payments as a fraction of net income will make the uniform VAT look progressive That is a distributionally neutral uniform VAT can be misleadingly labelled progressive or regressive if VAT payments are expressed as a proportion of net income For this reason analysis showing VAT payments as a proportion of household expenditure should
63. files are saved and then appended for data on regular expenditures weekly expenditures recorded in the expenditure diary and expenditure on education The relevant places can be found by searching for INSERT EXPENDITURE ADJUSTMENT FACTORS in the create_processed_data do file downloadable from the LATAX website It is important to note that national accounts provides information on only 12 categories of spending rather than the 60 30 formal 30 informal included in LATAX This means one must amalgamate the LATAX spending categories into larger categories by default exptype 1 to 10 for which National Accounts data is available and multiply by factors calculated for these larger categories For instance the code inserted at the end of the code creating information on education spending is gen exptype replace exptype 8 if clave gt E001 amp clave lt E007 replace gas_tri gas_tri 2 83 if exptype 1 replace gas_mon gas_mon 2 83 if exptype 1 replace gas_tri gas_tri 3 68 if exptype replace gas_mon gas_mon 3 68 if exptype 0 replace gas_tri gas_tri 11 87 if exptype 1 replace gas_mon gas_mon 11 87 if exptype 1 replace gas_tri gas_tri 1 34 if exptype 2 replace gas_mon gas_mon 1 34 if exptype replace gas_tri gas_tri 7 65 if exptype 3 replace gas_mon gas_mon 7 65 if exptype 37 replace gas_tri gas_tri 2 55 if exptype 4 replace gas_mon gas_mon 2 55 if exptype replace gas_tri gas_tri
64. folder in LATAX called Params_Data and place the parameters modules params_2008 params_2010 params_20101 e Create sub folders in LATAX called Input_Data and Output_Data in which the input datasets for use in the program and output datasets created by the program will be stored respectively Once this is done you should have a folder structure as illustrated in Figure 24 Figure 2 Recommended LATAX folder structure Organize v a Open Include in library Share with v Burn New folder Fr Favorites Name Date modified E Desktop J Do Files 04 12 2012 18 48 File folder B Downloads J Input Data 04 12 2012 18 48 File folder E Recent Places di Output_Data 04 12 2012 18 49 File folder A Libraries d Params_Data 04 12 2012 18 48 File folder a Libraries E Documents d Music Pictures B Videos jE Computer amp Local Disk C File folder a Output_Data Date modified 04 12 2012 18 49 Once the folders are created and the program used the next stage is to put the input datasets in to the Input_Data folder Instructions on how to create the input data can be found in Section 6 of this instruction manual Alternatively you can download datasets based on the 2008 version of the ENIGH from the LATAX website see above Last but not least load Stata and open up the main_module do and params_2008 do parameters file in the do file editor The program is now ready to be u
65. hat users know what each option parameter is used for B 1 main_module do This interface main module contains user edited instructions on the directories in which the simulation code input and output data and parameter code can be found the name and location of logfiles the names of input and output datasets the existing and reform systems to be used in the analysis and runs the tax simulator There are options for whether one wants to conduct the behavioral analysis In order to run LATAX the first thing necessary is to run main_module do Then the user must type in the following commands set_systems followed by the numerical name of 3 systems e g 2008 20100 20101 set_defaults simulator which runs the program calls oadprogs do The default options for running LATAX are contained in ain_module do and can be directly edited ain_module do Options can also be changed using interactive commands typed into the Stata command prompt Editable defaults in main_module do taxsim the directory of the LATAX do files in the directory that contains LATAX input data out the directory that LATAX output data will be written to param the directory containing the parameter definition do files logfiledecile the log file for cash and proportional tax changes by decile group logfiledecile1 the log file for proportion of change in tax revenues attributable to each decile logfilehhtype the log file for cash and propo
66. he assumption of less than full VAT and duties pass through and the associated direct effect of the change in VAT and duties on consumers are calculated using the assumed pass through rate Then new consumer prices are used to calculate pre tax prices and the amounts of VAT and duties paid under the reform systems The next stage is to allocate the part of the burden not borne directly to consumers to gross formal sector employment income and net capital income according to the ratio assumed in main_module do It is assumed that the amount of income of each individual with these sources is reduced by the same proportion We recalculate the amount of tax paid and net income given the changed gross income calculated in the previous stage by recalling set_taxbase and calc_dirtax When doing this we change the value of indic_incidence to 2 so that set_taxbase and calc_dirtax know that they are being called by indirect_incidence rather than load_progs The module then performs the same functions as do_household do and do_distanalysis do outputting to log files and an individual level data file logfiledecile_ind logfilehhtype_ind logfilerevenue_ind and outdata_ad_ind It should be noted that the tables in the decile and household log files show the proportional changes in net income and expenditure assuming fixed purchase quantities rather than changes in taxes paid This is because when pass through is less than complete the changes in taxes paid
67. hich is presently only the case for Mexico It has been designed in this way so that users do not have to edit the main program code even if they wish to make fairly major changes to the tax system e g introducing additional tax rates or the input data such as additional sources of income or expenditure categories Figure 1 shows the basic structure of the program We then describe each module in turn Figure 1 A graphical representation of LATAX Household Survey Data Data creation processing code to create necessary input variables from household surveys Further information available in Description of Data Creation for each country Individual File Household File Expenditure File Input Files Interface Module User sets Parameter Module User sets directories files parameters and parameters of the reform and models to use baseline systems Taxbase Module Calculates taxable Selected Input Files income based on system parameters and data Dirtax Module Calculates tax and net income based on rates and taxable income Indirtax Module Calculates expenditure taxes based on rates and input data Observation Output File Household Module Calculates household level taxes and changes in taxes Income Dist Output File Distribution Module Calculates average payments and changes by household groups HH Type Output File Incidence modules Repeats the above calculation when less than
68. hich no IVA is levied e Food and drinks and meals out on which IVA is levied e Alcoholic drinks and tobacco IVA and IEPS levied e Clothing and footwear IVA levied e Household goods services and communications IVA levied IEPS sometimes levied e Household goods services and communications no IVA levied e Transport and vehicle fuels IVA levied IEPS sometimes levied but not modelled e Public transport and other transport on which no IVA levied e Health and education goods no IVA levied e Health and personal goods and services IVA levied e Leisure and hotel services IVA sometimes levied e Other services IVA sometimes levied By aggregating goods in such a way our demand model is suitable for modelling the welfare impacts of changing the rate of IVA and imposing IVA on additional classes of goods However this level of aggregation means that we cannot model for instance substitution between different kinds of alcoholic beverage when the duties rates on different types of beverages change by different amounts Whilst this limits the number of questions the existing demand model can be used to assess we would argue that analysis of very detailed goods categories is best done using bespoke demand systems tailored to the question at hand Prices and associated expenditure weights used to calculate the prices of the aggregate commodities have been provided to us by the Bank of Mexico for 46 cities The Bank of Mexico also
69. hotels Non taxed household goods services and electronics Non taxed food and drinks Non taxed health and education Non taxed household goods services and electronics Leisure and hotels exp42 exp43 exp44 exp45 exp46 exp47 exp48 exp49 exp50 exp51 Informal Informal Informal Informal Informal Informal Informal Informal Informal Informal VAT zero rate VAT zero rate VAT taxed VAT taxed VAT taxed VAT taxed VAT taxed VAT taxed VAT taxed VAT taxed Health related goods and services Health related goods and services Food and drinks Food and drinks Household goods and services Personal goods and services including education and health Other non food goods and services Leisure services Transport goods and services Clothing footware and accesories Gastos Gastos Gastodiario Gastodiario Gastos Gastos Gastos Gastos Gastos Gastos Clave J009 J010 J014 JO20 J035 J037 J038 J042 J044 J059 J063 J064 T910 amp lug_com lt 4 Clave J061 amp lug_com lt 4 Clave A069 A071 A216 A217 amp lug_com lt 4 Clave A198 A202 A219 A222 A243 A247 T901 amp lug_com lt 4 Clave C001 C024 T903 GO08 G010 G020 G022 T909 F007 T911 1001 1026 K001 K044 L001 L022 T907 amp lug_com lt 4 Clave D001 D026 T904 E016 E018 E019 amp lug_com lt 4 Clave E009 E01
70. household goods services and electronics Leisure and hotels Non taxed health and education Taxed personal goods and services including taxed health education etc Non taxed food and drinks Taxed food drinks and food out Taxed household goods services communications and electronics Taxed personal goods and services including taxed health education etc exp18 exp19 exp20 exp21 exp22 exp23 exp24 exp25 exp26 exp27 exp28 exp29 Formal Formal Formal Formal Formal Formal Formal Formal Formal Formal Formal Formal VAT taxed VAT taxed VAT taxed VAT taxed VAT taxed VAT taxed VAT taxed VAT taxed VAT and duties VAT and duties VAT and duties VAT and duties Other non food goods and services Leisure services Transport goods and services Clothing footware and accesories Other non food goods and services Health related goods and services Petrol Telecoms Under 14 alcohol 14 20 alcohol Over 20 alcohol Beer Gastos Gastos Gastos Gastos Gastos Gastos Gastos Gastos Gastodiario Gastodiario Gastodiario Gastodiario Clave E009 E012 M005 NO08 N010 T915 amp lug com gt 3 Clave E023 E025 E027 E028 E030 E033 T912 LO23 L028 NO03 N005 amp lug_com gt 3 Clave F013 F017 T906 T913 M002 M004 M006 M018 amp lug_com gt 3 Clave H001
71. ile for the changes in spending patterns following tax reforms logfiledemand_revenue the log file for revenues from tax reforms allowing for demand effects NUMGOODSQUAIDS the number of good categories in the demand system 12 is default 41 NUMGOODSDEM the list of more disaggregated goods i e those contained in the indata_co file included in demand system a number of goods such as cats and other lumpy purchases may have to be excluded from the demand system goodslist 1 12 lists of disaggregated goods i e those contained in the indata_co file in each of the demand system categories goodslist13 A list of the disaggregated goods excluded from the demand system categslist 1 60 states the demand system category for each disaggregated good indir To perform analysis with less than complete IVA pass through set equal to 1 prop_prices The extent of pass through between 0 and 1 ptop_wages_ The extent to which taxes not passed on in prices are born by wages between 0 and 1 Four globals corresponding to logfiledecile logfilehhtype logfilerevenue and outdata_ad are used to define output files for analysis when different assumptions about IVA and duties pass through are made and should be changed for each scenario to avoid over writing files Changing options using interactive Stata commands If you do not want to change the default options in main_module do you can also change most but not all options
72. in logfiledecile1 in this case decileburden_practise as shown in Figure 16 Figure 16 Showing how the burden is borne across the distribution name lt unnamed gt log C LATAX Output_Data decileburden_practise txt log type text opened on 19 Dec 2012 11 41 13 PERCENTAGE OF ADDITIONAL REVENUE RAISED BY DECILES OF texp USING EQUIVALENCE SCALE es11 FOR REFORM SYSTEM 20100 Decile texp ISR Soc Sec IVA IEPS TOTAL pao Bohan eee ok eee renee eee eee esas ae eee ee eee eee eee Poorest 3 27 0 04 0 00 2 12 1 23 1 36 Decile 2 4 91 0 34 0 00 3 55 2 65 2 45 Decile 3 5 95 0 43 0 00 4 66 4 23 3 33 Decile 4 6 63 0 74 0 00 5 40 5 49 4 02 Decile 5 7 68 1 53 0 00 6 70 7 46 5 26 Decile 6 8 67 2 98 0 00 7 91 9 31 6 64 Decile 7 9 85 5 38 0 00 9 49 10 80 8 46 Decile 8 11 51 9 33 0 00 11 73 13 43 11 27 Decile 9 14 09 19 53 0 00 15 63 17 18 17 04 Richest 27 44 59 70 0 00 32 81 28 22 40 18 ee eee SI FE sir PSS A a ER ak ONS TE Average 10 00 10 00 0 00 10 00 10 00 10 00 The first column shows the fraction of total expenditure in this instance accounted for by each total expenditure decile group For instance the poorest 10 of households on this measure account for 3 27 of spending and the richest for 27 44 The next 4 columns show the fraction of gains or losses accruing to each decile group for changes in each of the 4 taxes I
73. ing the 2008 tax system as a baseline Open the params_2008 do file and immediately save it as params_20100 We will use this as the 2010 parameters file so as not to save over the pre made params_2010 file The first set of parameters OVAT OBVAT and ODUTIES refer to the VAT and duty rates as they were when the survey data being used was collected The default values are the 2008 rates so if you are using 2008 data which is the default presently no changes need to be made here If on the other hand you are using a more up to date dataset 2010 or 2012 which will be released in 2013 you will need to update these values to the correct rates for that year The first thing that does need changing is the actual VAT rates stored in the VAT and BVAT scalars Initially 14 goods categories are either exempt from VAT or subject to a zero rate so that VAT and BVAT are equal to 0 The next 16 goods are subject to either a VAT rate of 15 VAT 0 15 or 10 for certain border areas BVAT 0 1 The remaining 30 goods represent informally purchased goods and services for which we assume no VAT is paid The 2010 reforms raised the rate of VAT from 15 to 16 and the special rate of VAT applying in border areas from 10 to 11 To make this change replace VAT 0 15 with VAT 0 16 for 15 to 30 and BVAT 0 1 with BVAT 0 11 for 15 to 30 Figure 3 shows what the parameters should look like after changes have been made The 2010 reforms als
74. ional impact of tax reforms The literature on tax reform analys has noted that whether income or expenditure is used to rank households and to calculate proportional gains losses can have a major impact on the assessed distributional impacts Abramovsky et al 2011 IFS et al 2011 For instance the VAT systems operating in most countries look regressive when VAT paid is expressed as a fraction of income over the income distribution but slightly progressive when VAT paid is expressed as a fraction of expenditure over the income or expenditure distribution This would appear to present significant difficulties to the analyst results are highly sensitive to how one decides to calculate the proportional burden of a tax But can economic reasoning guide us to the right way to analyse reforms First up itis not not simply a question of using income or expenditure There are two distinct questions which may have different answers e How should one rank households as rich or poor e How should one assess the proportional change in tax payments resulting from a reform As is explained in Box 5 1 what is really needed to determine reform progressivity is lifetime income and expenditure or a good proxy for this which is not available 25 The answers to both are strongly linked to the long term or life time distributional impact of a tax change but in different ways First how should households be ranked The issue
75. irect tax simulator This is done to calculate total expenditure on goods included in the demand system which together with prices read in from indata_prices and the coefficients from the demand system read in from indata_demand is used to calculate indirect utility and the estimated shares for each of the 12 demand system goods categories for each household The price effects of tax reforms are then calculated and are used to estimate the shares for each of the 12 goods categories following the reforms The expenditure function is then calculated to calculate the compensating variation CV for each reform which after adding on the change in taxes paid on goods which we cannot include in our demand system is our measure of the welfare effect of the tax reforms accounting for substitution possibilities We also calculate the change in expenditure required to obtain the same level of utility if substitution were not possible so that we can evaluate the extent to which substitution possibilities ameliorate the impact of tax changes The changes in consumer welfare by decile group and household type are calculated and outputted in logfiledemand_welfare the changes in expenditure shares are outputted in logfiledemand_spending and the changes in revenues are outputted in logfiledemand_revenue 3 Installation requirements and procedures LATAX is written in Stata and has been successfully used with versions 10 11 and 12 A key requirement for u
76. is is a practise exercise Never change the final control parameter indic_incidence 1 The final step is to run the program To do this e Click the execute do button on the right hand side of the do file editor toolbar This runs the code e Then in the main Stata command line type without quotations set_systems followed by the numbers of the baseline and two reform systems In this instance set_systems 2008 20100 20101 e Then type set_defaults e Then type simulator The program is now running Once it has finished you can look at outputs see Section 5 Programming LATAX using Stata command line programs If you decide to set up LATAX using Stata command line commands rather than by amending the main_module do file the first stage is to click the execute do button on the right hand side of the do file editor toolbar ensuring the main_module do file is the active do file displayed on screen before doing this Then in the main Stata command line type without quotations set_systems followed by the numbers of the baseline and two reform systems In this instance set_systems 2008 2010 20101 followed by set_defaults You are now ready to start telling LATAX directory locations filenames and selecting program options A full list of the optional command line commands can be found in appendix B Here we describe those needed to set the program up for the practise exercise as
77. iture decile groups The global disttype defined in main_module do tells this module whether the full set of distributional analysis should be conducted disttype 1 or only a partial analysis disttype 2 Average cash gains losses per household by decile group are calculated as minus the sum of the change in tax payments in each particular decile group divided by the number of households in each particular decile group The average proportional gains losses per household by decile group are calculated as minus the sum of the change in tax payments in each particular decile group divided by the total income expenditure of households in each 6 particular decile group In addition the proportion of the increase or decrease in revenue attributable to each decile group is also calculated for each tax and in total and compared to the proportion of total income expenditure attributable to each decile group The results are displayed in tables which are saved in log files The process is repeated for gains losses by household type indirect_incidence do This program allows one to carry out analyses assuming that indirect taxes are partly incident on wages and profits instead of fully on prices The program first loads output data from the indirect tax calculations performed in calc_indirtax do and input data from the files indata_ad and indata_hh Calculations then proceed as follows First post tax reform consumer prices under t
78. l i homogeneity n i 1 symmetry Yij Vii Negativity cannot be imposed in such a manner but the estimated Slutsky matrix can be tested to see if it satisfies this criterion This paper allows for household demographics to affect demands in a fully theoretically consistent manner Demographics enter as taste shifters in the share equations and to maintain integrability they are therefore part of a terms in Ina p K 1 Ina p a fa DI cust 5 Yy mP p i k 1 i j Which gives us the following new adding up conditions that supersede Xf aj 1 a 1 Qik 0 n n i 1 i 1 Calculating the Welfare Impact of Price Changes Having estimated a fully specified demand system one can estimate the impact of price changes on consumer welfare using the associated expenditure functions An attractive measure of the welfare impact is the compensating variation CV the change in income a household would require in order to make them indifferent between the original price vector with the original income and the new price vector This is calculated as CV E u p E u p where u is the original value of the utility index p is the initial price vector pt is the new price vector and E u p y 0 1 is 1 E u p o Ma D r AD 61 and where In u can be calculated using the indirect utility function Price and total expenditure elasticities are derived and presented in Banks et al 1997
79. labor labor practise txt 2 CONSUMPTION MODEL do you want to run consumption model 1 Yes 2 No DEFAULT SET TO 2 NO global consmod 1 do you want to reduce spending in response to changes in dire global inc aff spend 1 Others global indata _ demand demand params dta global indata_prices demand prices dta global logfiledemand welfare decile demand practise txt global logfiledemand spending spending demand practise txt Global logfiledemand revenue revenue demand practise txt Next are options which tell LATAX whether to perform behavioural analysis The global labmod determines whether labor supply modelling using assumed participation and hours earnings elasticities is carried out The default is for it not to be 2 Amend this to 1 so that such modelling is done logfilelabor is the name of the file that records the outputted labor supply effects For this practise change this to labour_practise txt 16 The global consmod determines whether consumer demand modelling is carried out The default is for it not to be 2 Amend this to 1 so that such modelling is done inc_aff_spend tells LATAX whether or not to assume that changes in net income following changes in income tax or social security contributions feed into changes in consumption The default is that they do not 2 changing this global to 1 means that they do the percentage change in total expenditure is assu
80. lds purchasing certain items infrequently especially larger durable goods such as motor vehicles or new kitchens but also food if they bulk purchase Excluding durable goods from the measure of expenditure removes much of this problem but introduces a new one you may rank households incorrectly if they devote different proportions of their budgets to durable goods It is therefore not clear whether expenditure represents a better measure of a household s living standards than income both are volatile and furthermore both suffer significant measurement error in surveys For this reason it is worthwhile conducting analysis ranking households both according to their position in the income distribution and in the expenditure distribution Second how should we assess proportional gains or losses The best way to understand why gains or losses should be expressed as a fraction of income for direct taxes and as a fraction of expenditure for indirect taxes is to use some hypothetical tax changes As this issue most often comes up when analysing changes to indirect taxes we use examples for VAT that demonstrate how presenting results as a fraction of expenditure is most appropriate but analogous examples can be used to show how income should be used for direct changes Consider the case of a uniform VAT on all goods and services Over a lifetime if lifetime income and lifetime expenditure are equal this can be clearly seen as distributionally n
81. low one to conduct distributional analysis for the entire population and calculate aggregate revenue effects of tax reforms using the sample of individuals households for which data is available LATAX is based on a modular approach e User amends interface module which contains settings on the type of analysis to be performed and certain assumptions to be made e And the parameter modules which include tax rates thresholds etc under the base and reform systems it can handle one base system and two reform systems e Tax calculations and distributional analysis modules are not system specific and need not be amended for a large number of reforms e This means it is easy to do many actual and counterfactual reforms with limited knowledge of how the program calculates taxes A tax micro simulation model can be static or dynamic and can include no behavioral response assumed behavioral response or estimated behavioral response 7 In existing applications of LATAX for Mexico only income taxes on labour income have been simulated due to low quality information capital income in the household survey used 2 It outputs a household level dataset with incomes expenditures and tax payments and log files with summary distributional and revenue figures It is hoped that the model will be useful to academics civil society organisations international bodies such as the World Bank and civil servants In order to help such use
82. med to be the same as the percentage change in net income The next set of globals determines input files for the consumer demand modelling indata_demand contains the demand system parameters and indata_prices contains data on the prices of different goods and the logfiles used for outputs beginning with logfiledemand_ In this practise exercise add _practise to the end of each logfile file name If using the default demand system and input data included with LATAX for Mexico do not change the input datasets or the remaining consumer demand globals NUMGOODSQUAIDS to categslist60 Appendix D describes what to do if you wish to estimate a new demand system and incorporate that into the model The final behavioural model is that which allows firms to pass less than 100 of changes in indirect taxes IVA and IEPS through to consumer prices see Figure 9 The default is that this option is turned off indir 2 but in for this practise exercise let us turn it on indir 1 Figure 9 Amended control variables labor supply and consumer demand options 3 INDIRECT TAX INCIDENCE do you want to estimate distributional effects with alternat DEFAULT SET TO 2 NO global indir 1 proportion of pass through global prop prices 0 75 proportion not passed through borne by wages global prop wages 0 75 output data for the different pass through of indirect tax global outdata_ad ind out_adlevel_ind p75 _w75_ practise
83. n this instance around 0 04 of the extra income tax revenue is estimated to be raised from the poorest 10 of households and 59 7 from the richest 10 Note that positive numbers in this table mean that households in a decile group are sharing at least a bit in the overall burden of a tax raise or gaining at least a bit from a tax cut a negative number means that whilst households on average are losing gaining households in that decile group are gaining losing The last column shows the fraction of gains or losses accruing to each decile group for the tax reforms as a whole 28 The last log file for the no behavioural response analysis is that declared in logfilehhtype in this case hhtype_practise as shown in Figure 16 This shows the impact of reforms in cash and proportional i e as a percentage of income and expenditure terms for different types of households as defined in LATAX The format is effectively the same as for logfiledecile The household types are Couple Single Couple children lt 18 Single children lt 18 Pensioner single Pensioner couple Other family no lt 18 Other family with lt 18 Multi family no lt 18 Household containing one couple with no child aged under 18 Household containing a single adult with no child aged under 18 Household containing one couple and one or more child aged under 18 Household containing a single adult and one or more child aged under 18
84. ndard errors are calculated using bootstrapping with 500 iterations Rather than draw the bootstrap samples in an unrestricted manner we take into account that we use variation in prices across city regions clusters and draw with replacement from within clusters as opposed to from the entire sample D 3 Data description This section provides further information about the data used in the estimation of the demand system These data should be the same as the one used in the MEXTAX micro simulator To generate these data the program create_demand_data_dems do has been used The demand system is estimated in the program quaids_estimation_dems do which is downloadable from the LATAX website see above In order to ensure that the model can be feasibly estimated it is necessary to aggregate the very detailed expenditure breakdowns in ENIGH into a significantly smaller number of aggregate commodity groups These are designed to ensure both that the groups make sense as functional 1 That is we regress Inx and Inx on the prices and demographic variables included in our demand system and on the log of household monetary income and the square of the log of household monetary income and include cubic terms of the residuals from these regressions in our demand system equations 62 product groups but also to allow for substitution between goods treated differently by the indirect tax system The 12 categories chosen are1 e Food and drinks on w
85. ng patterns If inc_aff_spend 2 instead then only the impact of changes in indirect taxes on spending patterns would be modelled Figure 19 shows the output of this logfile Figure 19 The impact of reforms on spending patterns IMPACT OF REFORMS ON SPENDING PATTERNS Commodity Group Original Reform 20100 Reform 20101 ewe we we we we we we we ww we we ee ee eee 1 26 886 26 898 27 176 2 12 905 12 897 12 832 3 0 642 0 649 0 641 4 7 168 7 177 7 160 5 21 608 21 567 21 495 6 1 633 1 638 1 641 Fj 7 318 7 288 7 216 8 6 344 6 378 6 394 9 3 195 3 193 3 188 10 7 580 7 597 7 590 11 4 087 4 078 4 033 12 0 634 0 640 0 634 ee ee ew ew ew ee ee ee ee ee eee A list of commodity groups can be found in Appendix D The first column shows the predicted fraction of aggregate consumer spending going towards each commodity group prior to the reforms The second column shows the predicted fraction of aggregate consumer spending going towards each commodity group after the first set of modelled reforms system 20100 with the last column showing the same for the second set of modelled reforms 31 The last logfile produced by this module shows revenue figures before and after consumer demand responses are accounted for declared in logfiledemand_revenue in this instance revenue_demand_practise txt The first table shows the revenue
86. nship to head of household all of which can be created from the pobla ENIGH file 34 The variable empstat states whether an individual is working or not and is based on the ENIGH variable trabajo which records whether someone worked in the last month or not The variable formal_w is an indicator of whether a person is employed formally and therefore assumed to pay tax and social security contributions on all income on which such payments are liable or informally and therefore is assumed to evade all such payments The default is for this to be based on ENIGH variables which records whether an individual has access to healthcare provided by a social security scheme and whether that access is via their own employment This was the best proxy that could be found after significant search so it is not advised to change this without serious thought Finally the remaining variables record the gross income an individual has from each source The default is for there to be 20 sources incl ince20 the first 10 of which record formal sector income and the next 10 which record informal sector income incomes are allocated to each source in create_grossincome do with formality based on the formal_w variable The user is able to amend the number of sources of income when creating their own input dataset although doing this will require changes to the parameters files to change the number of sources and the tax treatment of each source An import
87. o saw an increase in certain duty IEPS rates and the introduction of a 3 tax on telecommunications services The DUTIES scalars store the duty rates expressed as a percentage of the pre tax price 6 By default the goods subject to duties are listed as goods 25 30 To simulate the 2010 reforms make the following changes e DUTIES25 0 03 from 0 00 the 3 telecoms tax e DUTIES26 0 265 from 0 25 the increase in tax on wine and alcohol lt 20 alcohol e DUTIES27 0 3 from 0 2 the increase in gaming lottery tax in principle in reality duties in Mexico are a mix of specific duties per unit or by volume and ad valorem percentage duties Because of poor quality information in the ENIGH survey on quantities purchased we used average ad valorem rates calculated by CEFP It would be possible to amend the tax simulator to allow for ad valorem and specific duty rates by amending the consumption input dataset to include quantities and the indirect tax parameters and simulation code accordingly 10 e DUTIES28 0 53 from 0 5 the increase in taxes on spirits gt 20 alcohol e DUTIES29 0 265 from 0 25 the increase in tax on beer e DUTIES30 1 64 from 1 60 the increase in tax on cigarettes Figure 4 shows the duty rates following these changes Figure 3 VAT rates following the 2010 tax reforms REKKKKHKKKHRKKHKKHREEEE NEW VAT 8 8 ee eR KKK KKK KEEE scalar VATi scalar VAT2 scalar VAT3 scalar VAT4 scal
88. only e And using different equivalence scales to adjust incomes and expenditures for household size and composition when ranking households into decile groups Thus for instance the first two tables show the impact of the first set of reforms in this case system 20100 in cash terms the first table in Figure 14 and as a proportion of total expenditure texp the second table in Figure 14 for deciles of the total expenditure texp distribution when households are ranked using an equivalence scale whereby expenditure and income is divided by the number of household members es11 This is what the heading IMPACT OF REFORMS AS A PROPORTION OF texp USING DECILES OF texp USING EQUIVALENCE SCALE es11 FOR REFORM SYSTEM 20100 means Figure 14 Interpreting the decile log file name lt unnamed gt log C LATAX Output_Data decile_practise txt log type text opened on 19 Dec 2012 11 39 18 impact OF REFORMS IN CASH TERMS USING DECILES OF texp USING EQUIVALENCE SCALE esil FOR REFORM SYSTEM 20100 Decile texp ISR Soc sec IVA IEPS TOTAL E ene R t I DEE E EE ERER E EE EE EE E EE E EEA EAS EA ES Poorest 38 626 1 is 86 14 101 Decile 2 57 940 8 0 145 30 183 Decile 3 70 275 10 0 190 49 249 Decile 4 78 243 17 0 220 63 300 Decile 5 90 825 34 0 273 86 394 Decile 6 102 458 67 ie 323 107 496 Decile 7 116 323 121 0
89. oss income can be found in Section 6 2 create_LATAX_input_data do is the final do file which creates the household adult and expenditure files used in LATAX itself from the files created by the previous do files By default these outputted files are called household dta adult dta and consumption_monetary dta which can then be renamed for use in the simulator e g as Input_Data_hh for the household file in the default implementation of LATAX Section 6 1 provides more detail on what is required of the household input file section 6 2 covers the adult input file and section 6 3 covers the expenditure input file Section 6 4 provides a brief explanation on how to amend the input data to account for under reporting of income and expenditure in the ENIGH survey 6 1 The household input file The household input file contains information on household composition and type calculated equivalence scales a sample weight and information sampling a geographic identifier and information on household net total monetary and non monetary income and expenditure A full list of variables can be found in table C1 of Appendix C with those in bold required and those in falics optional We now describe the main required variables 33 The first set of variables are the household identifiers By default these are folioviv and foliohog which identify the address and the household within that address respectively These are taken
90. oss lt 60 replace partelast partelast 3 if percgross gt 20 amp percgross lt 40 replace partelast partelast 4 if percgross gt 0 amp percgross lt 20 replace partelast partelast 2 if menores gt 1 amp sex 2 5 Interpreting outputs of model distribution revenue behavior Once the program has finished running you will see end of do file at the bottom of the Stata output window you can then start looking at the outputs of the model These outputs will consist of a number of Stata dta files and text log files When looking for summary distributional revenue and behavioral measures of reforms the log files provide the necessary information therefore they are the focus in this section Information on the variables contained in the output files can be found in Appendix C 5 1 Main log files from the no behavioural response simulations The outputs of the simulation will be stored according to the directory out and filenames chosen when setting up the LATAX simulation Four log files are produced for the no behavioural response simulations Then first records the revenue effects of the reform and is stored in the filename declared in logfilerevenue in this case revenue_practise Figure 13 shows the format of this output Figure 13 The revenue log file Ej revenue_practise Notepad File Edit Format View Help name lt unnamed gt Jog C LATAxX output_Data revenue_practise txt log type text opened on 1
91. p75_w75_practise txt 32 logfilerevenue_ind in this case revenue_ind_p75_w75_practise txt 6 Input data This section describes briefly how to create the input datasets required for use in LATAX for Mexico The model requires three input files a household demographics file nousehold dta an individual demographics and income file adult dta and a household expenditure file consumption_monetary dta Data creation files used to create our 2008 input datasets from the 2008 ENIGH can be found at the LATAX website see above for URL these can be used as a basis for developing input files with different variables or for different years There are four do files used to create the input dataset and they need to be run in the following order create_processed_data do this file processes the raw ENIGH files to create files containing information on demographics expenditure by category and household net income by source It also creates variables identifying formal sector workers and splits expenditure into formal sector and informal sector purchases create_gtossincome do this file converts net income by source into a gross income by source using information on formality status from create_processed_data do to determine whether tax has been paid by a particular individual It also calls clave_ingresos do in order to label data correctly Further information on how this do file works and what is required to create gr
92. provided data on the city whose prices are used for each municipality with a population of greater than 15 000 that is included in the ENIGH The Bank determines these linkages using distance population size and other characteristics Links between the cities and municipalities of less than 15 000 people are not made by the Bank and are instead computed by the authors using travel time according to Google maps This simple method was chosen to ensure ease of replication by other researchers and for future and past waves of ENIGH The prices of the aggregated commodities are calculated as weighted arithmetic averages of the prices of the individual goods making up the commodity Arithmetic as opposed to geometric averages termed Stone prices are used because geometric averages assume within group own price elasticities of 1 and compensating cross price elasticities which would mean that the welfare costs of changes in indirect taxes would be lower than when not allowing for behavioral response by assumption rather than because of the demand system estimates of the potential for substitution between different aggregate commodities We model changes in indirect taxes as increases in the prices of the aggregate commodities When a commodity contains goods which are seeing differential proportional increases in prices due to changes in tax for instance the alcohol and tobacco group where different 1 A full description of the products in each catego
93. ptions for different groups cluttering up the main_program interface module these have been hard coded into labor do Users who wish to change these for instance who want to assume labor supply is more or less responsive than the assumptions built into LATAX as standard which are broadly on the high side for developed countries but were felt to be a reasonable central guess where responsiveness could be greater due to the ability to work informally will therefore need to open labor do do file and make changes directly there This is easy to do if one wants to allow elasticities to vary by basic features such as sex number of children age position in the earnings distribution etc as these variables are already defined in the standard LATAX input datasets or calculated in the labor do program The labor supply module uses a set of elasticities for the extensive participation employment decision partelast and the intensive hours decision hourselast A higher elasticity implies more behavioral response to a change in tax rates and for instance a lower yield from tax increases or a smaller cost for tax cuts To edit the elasticities the following blocks of code need to be amended gen hourselast 0 05 replace hourselast 0 1 if sex 2 amp menores gt 1 replace hourselast 0 1 if percgross gt 90 replace houtselast 0 2 if percgross 100 21 gen partelast 0 05 replace partelast partelast 2 if percgross gt 40 amp percgr
94. raised from the changes to indirect taxes after consumer demand effects are allowed for with the second column showing the revenue raised if one assumes there is no behavioural response Figure 20 shows this Figure 20 Impact of consumer demand changes on revenues from indirect tax changes Revenue raised from the tax reforms after behavior Tax Reform System 20100 Reform System 20101 eee eee ee eo ewww p see IVA 6 753 457 664 30 980 372 480 IEPS 2 778 789 888 3 584 357 888 wee a ee ee eo ewww e ee Revenue raised from the tax reforms before behavior Tax Reform System 20100 Reform System 20101 m l l l l l l l l IVA 10 888 586 240 38 020 403 200 IEPS 3 064 007 424 4 083 248 640 aa ae ee M M M D eo oon eee 5 4 The less than full pass through of indirect tax changes log files Three output files are produced that show the results of the simulations when less than full pass through of indirect taxes are assumed These follow the same format as the outputs for the regular no behavioral response simulations and users are invited to examine that section for understanding how the outputs are arranged It is worth mentioning however a number of conceptual methodological issues here First that the analysis produced by this module is very much an approximation In particular
95. rds some items on a weekly basis others on a monthly and others on longer time period must be converted to an annual equivalent The final variable in this file is the region variable region which allows for VAT rates to differ for certain households By default region is set equal to 0 for all households but setting region 1 means that a household will face the BVAT rates of VAT defined in the parameters files rather than the standard VAT rate This allows one in principle to model the reduced rate of VAT in certain border areas of Mexico if municipality boundaries align well with these areas for instance 6 4 Accounting for missing income and expenditure The ENIGH surveys significantly under record expenditure particularly on items like alcohol and tobacco and incomes particularly capital income relative to those recorded in the Mexican National Accounts For instance as reported in Abramovsky et al 2011 in order to match National Accounts figures monetary employment income would need to be multiplied by a factor of 1 62 1 1 on average self employment income by 2 52 1 1 capital and rental income by 26 04 1 1 and transfer income by 1 30 the taxable income source factors ate divided by 1 1 as the average tax rate is assumed to be around 10 The default input dataset does not make any adjustments to account for this under reporting but we do provide input datasets that account for under recording using fixed factors for
96. rs get the most out of LATAX this manual explains how to set up and use the model using a worked example of simulating the distributional and revenue impacts of the 2010 Mexican tax reforms Details of the reforms being analysed can be found in Appendix A 2 Overview of program structure LATAX is written in Stata code and is designed so that users do not need to edit the main simulation code Instead basic instructions such as the names of input and output datasets which tax systems to use and whether to run behavioral response modules can be input either interactively via the Stata command line or input into a main interface module and the parameters of the baseline and reform tax systems changed by amending system parameters modules which define the basic structure and tax rates Based on the input data and the user defined tax parameters separate modules then calculate indirect tax payments the direct tax base and direct tax payments before calculating the revenue effects of the reforms and the impact of the tax changes across the income expenditure distributions and by household types Separate modules can be turned on and off according to need to allow for less than full pass through of changes in indirect taxes to changes in consumer prices as well as to model labor supply or more correctly taxable income and consumer demand responses to tax changes if a demand system has been estimated for the country LATAX is being used for w
97. rtional tax changes by household type logfilerevenue the log file for revenue changes from tax reforms indata_hh household data input file indata_ad individual and incomes data input file 40 indata_co expenditure data input file outdata_ad_ individual level output file oudtdata_hh household level output file sys1 baseline system number e g 2010 sys2 reform system 1 number e g 2011 sys3 reform system 2 number e g 2012 disttype set equal to 1 to perform analysis with all equivalence scales 2 for just 100 80 50 scale netincometype set equal to 1 to use net income calculated using LATAX 2 to use a net income measure directly recorded in the input data global identlist the household identifier variable or set of variables There are then a set of globals which are used to determine the type of behavioral analysis to be performed labmod To run labor supply analysis set equal to 1 2 otherwise logfilelabor the log file for output of labor supply module consmod To run the QUAIDs demand analysis set equal to 1 inc_aff_spend Set equal to 1 for demand modelling to account for changes in net income due to direct tax changes 2 to exclude this effect 2 is default indata_demand input file containing parameters of demand system indata_prices input file containing prices logfiledemand_welfare the log file for the welfare effects of tax reforms logfiledemand_spending the log f
98. ry can be found in table C 1 in Appendix C 1 For instance if one wanted to estimate the impact of differential taxation of forms of alcohol and changes in alcohol taxation one may want as categories the various forms of alcohol tobacco non alcoholic drinks food out food in other leisure and other goods and services 63 changes may be made to the duty rates the increase in the price of that commodity is the arithmetic weighted average of price changes of the goods within the category averaged across the entire population If the within group composition differs systematically across the population for instance poorer households consuming alcohol in the form of beer and richer households in the form of wine or the extent of tax evasion varies across the population this estimated change in prices might not reflect well the change in prices for the types of goods within that group that particular households face This is a limitation of the demand modelling more so for the analysis of changes in IEPS than for changes in IVA and one that should be improved upon in future analysis of consumer responses to expenditure taxation in Mexico Our measure of the welfare cost of indirect tax changes is called the compensating variation CV that is the amount of income required to allow a consumer to obtain the same level of utility when prices increase due to increases in indirect tax once one allows for the ability to substitute bet
99. s clave P001 annual pobla08 P002 P003 income in trabajos P006 P011 Mexican P015 2008 inc12 informal Gross Overtime ingresos clave P004 annual pobla08 income in trabajos Mexican 49 2008 inc13 informal Gross End of the year bonus ingresos clave P009 annual pobla08 income in trabajos Mexican 2008 inc14 informal Gross Incentives rewards and prizes ingresos clave P005 annual pobla08 income in trabajos Mexican 2008 inc15 informal Gross Holiday bonuses and cash ingresos clave P007 annual allowances pobla08 income in trabajos Mexican 2008 inc16 informal Gross Profit sharing ingresos clave P008 annual pobla08 P019 income in trabajos Mexican 2008 inc17 informal Gross Pensions ingresos clave P032 annual pobla08 income in trabajos Mexican 2008 inc18 informal Gross Other labor income ingresos Clave P013 annual pobla08 P017 P018 net income in trabajos of P019 Mexican 2008 inc19 informal Net annual Capital Clave P016 income in P023 P031 Mexican P047 P065 2008 inc20 informal Net annual Other income inc_semp income in clave P067 Mexican P080 inc_tran clave P033 P045 P066 inc_other clave P046 inc_oemp clave P020 P022 P063 P064 50 Table C 3 Consumption file All expenditure categories exp are annual figures in Mexican 2008 Variables folioviv foliohog exp1 exp2 exp3 exp4 exp5 exp6
100. s and bequests is hit infinitely hard by VAT Including bequests and gifts in the lifetime resources of the recipient makes subtracting them from the resources of the giver attractive to avoid the double counting of gifts and bequests Adding and subtracting gifts and bequests when calculating lifetime resources in this manner means a uniform VAT would be found to be a constant fraction of both lifetime resources income and lifetime expenditure i e it would be distributionally neutral as in the case with no gifts and bequests In paying careful attention to the proportional impact of a set of reforms one must not forget to pay heed to how the reforms affect households in cash as well For instance a reduced rate of VAT for food might be progressive in the sense that it represents a larger gain as a fraction of spending for poorer households than richer ones on average This is because food represents a larger fraction of the budgets of poorer households But richer households spend more on food in absolute cash terms and hence would gain the most from such a policy in cash terms suggesting it is not a particularly efficient way to redistribute Thus a more complete picture of the reforms can be obtained by considering both the cash and proportional tables included in the logfiledecile log file Additional information on the percentage of overall gains or losses from reforms that are attributable to each decile group can be found in the log file declared
101. s to change the measure of total spending used in the demand system the demand system currently uses monthly spending equivalised using the eqsc_8050 scale additional The default variables are named child no of children adult number of adults sex sex of household head empstat employment status of household head educlow household head has low education educid household head has medium education v1 v6 polynomials of the errors of first stage regression of total expenditure on total income central north_interior north_coastal west east south and south_east regional identifiers 70 changes will need to be made to the quaids do file In particular one will need to change the global dems and search for eqsc_8050 12 in quaids do and edit appropriately Changing the demand model from the QUAIDS form would require a comprehensive edit of quaids do that is beyond the scope of this manual and should only be considered by advanced users Finally it is worth noting some good practises when developing a demand system for use with LATAX or any tax microsimulation model e Ensure that there is sufficient exogenous variation in prices to estimate the demand system This is unlikely to be the case if demand system categories are very narrow and there are few observations e Align demand system categories with tax system treatments in baseline and reform systems so that one can properly model the substitution that
102. sed Section 4 explains how to use it using our worked example If you are planning to use LATAX for analysis in multiple countries it is recommended that you create sub folders in the LATAX for each country the name should contain no spaces e g ElSalv for El Salvador Then place the Input_Data Output_Data and Params_Data folders and relevant data and do files for each country in that country sub folder 4 Running the model interface module and parameters This section uses a worked example simulating the 2010 Mexican tax reforms to explain how to use the LATAX program It begins with Stata open and the programs main_module do and params_2008 do open in the do file editor see Section 3 4 1 Setting up the parameters files The first step is to get the systems parameters files ready for use LATAX always requires a baseline system in this practise exercise the pre reform 2008 system params_2008 and two reforms systems here the implemented and initially proposed systems for 2010 params_2010 and params_20101 respectively and parameters files need to be created for each of these systems The parameters files contain information on the various tax rates thresholds exemptions and allowances required to model the different parts of the tax system A description of every parameter can be found in appendix B here we focus on implementing the changes to the parameters required to model the 2010 tax reforms us
103. sing the program is therefore to have Stata installed contract your system administrator or see www stata com for information on how to obtain and install Stata if you do not already have access to it With Stata installed download the LATAX package from the LATAX website The downloaded file is a zip file the do files making up the LATAX program need to be extracted and placed in folders from they can subsequently be called when using the program There are two types of do files program modules main_module do loadprogs do calc_indirtax do set_taxbase do calc_dirtax do do_household do do_distanalysis do indirect_incidence do labor do and quaids do and 3 example parameters modules which are used in the worked example in this user manual and can be used as the basis for your parameters modules params_2008 params_2010 params_20101 Where the different types of files are placed is to a large extent up to you the program will work if they are all placed in one folder or the program files and parameters files are placed in separate folders However it is recommended initially to use the following convention e Create a folder called LATAX e Create a sub folder in LATAX called Do_Files and place the program modules main_module do loadprogs do calc_indirtax do set_taxbase do calc_dirtax do do_household do do_distanalysis do indirect_incidence do labor do and quaids do in that folder e Create a sub
104. the names of scalars are not changed although of course the values can be For indirect taxes the user needs to define the following scalars NUMGOODS The number of expenditure categories in the input data OVAT 1 NUMGOODS The standard VAT rate applicable to a particular category in the year the input data was collected This should not be changed when changing the parameters of a reform or baseline system but only when the data used is from a different year e g 2010 OBVAT 1 NUMGOODS __ If different VAT rates are applied in different parts of the country this allows a second rate of VAT to be set It records the second rate applicable to a particular category in the year the input data was collected This should not be changed when changing the parameters of a reform or baseline system but only when the data used is from a different year 44 ODUTIES 1 NUMGOODS The duties rate applicable to a particular category in the year the input data was collected This should not be changed when changing the parameters of a reform or baseline system but only when the data used is from a different year VAT 1 NUMGOODS The standard VAT rate applicable to a particular category in the tax system under consideration This should be changed when changing the parameters of a reform or baseline system BVAT 1 NUMGOODS If different VAT rates are applied in different parts of the country this allows a second rate of VAT to be set rate
105. total per household set_taxbase do First depending on the value of the global indic_incidence defined in the main_module either the standard input data is loaded indic_incidence 1 or the data adjusted for less than full pass through indic_incidence 2 or 3 Then using the amount of each income from each source and the exemptions EXEMPT PEXEMPT MPEXEMPT SSEXEMPT SSPEXEMPT and SSMPEXEMPT the taxbase for income tax and social security contributions is calculated Sections allowing for deductions of certain expenses such as medical insurance are currently commented out as the inability to model income tax on capital and self employment income partly due to poor data means that one cannot calculate the maximum amount deductable which is typically a fraction of taxable income calc_dirtax do The first part of the program calculates social security contributions using the social security bands and rates LSOCSECX_Y SOCSECX_Y to calculate fixed quotas for social security contributions that is the amount of contributions paid on income up to the start of each band This means that when calculating the amount of social security contributions paid this can be added on to the amount paid on income within the band an individual finds themselves in simplifying calculations considerably The same process is then done for income tax payments after deducting the amount of social security contributions paid from the income tax base if DE
106. ture monetary and non monetary in Mexican monexp Annual monetary expenditure in Mexican nmonexp Annual non monetary expenditure in Mexican moninc_e Annual net total current income monetary non monetary in Mexican moninc Annual net monetary current income in Mexican totinc Annual net employment monetary current income in Mexican hh_change_inctax sys2 sys3 Change in income tax paid between base and each reform system hh_change_socsec sys2 sys3 Change in social security contributions paid between base and each reform system hh_change_ieps sys2 sys3 Change in IEPS paid between base and each reform system 58 hh_change_iva sys2 sys3 Change in IVA paid between base and each reform system hh_change_tot sys2 sys3 Change in total tax paid between base and each reform system hhnet sys1 sys2 sys3 Household net income under each system 59 Appendix D This appendix presents the Quadratic Almost Ideal Demand System QUAIDS used in the quaids do module in detail First the theoretical basis of the model is explained before moving on to discuss how we estimated the model D 1 Assumptions and welfare impacts The impact of tax reforms on consumer spending patterns and welfare can be simulating using a built in Quadratic Almost Ideal Demand System QUAIDS This is a useful demand system that allows the share of each type of good in total expenditure to vary in a flexible way with goods
107. ucation 0 otherwise Households in municipalities associated with the following cities Cuernavaca Puebla Queretaro Guadalajara Aguascalientes San Luis Petosi Tlaxcala Morelia Jacona Iguala Leon Tepatitlan Tolouca Tulancingo Cortazar Households in municipalities associated with the following cities Jimenez Cuidad Juarez Chihuahua Monclova Monterrey Torreon Durango Fresnillo Cuidad Acuna Households in municipalities associated with the following cities La Paz Mexicali Matamoros Hermosillo Huatabampo Tijuana Culiacan Households in municipalities associated with the following cities Tepic Acapulco Colima Households in municipalities associated with the following cities Cordoba Verazruz San Andres Tuxtla Tampico Households in municipalities associated with the following cities Oaxaca Tehuantepec Tapachula Villahermosa Households in municipalities associated with the following cities Campeche Chetumal Merida 67 No prices information for other services ENIGH codes N006 N007 NO09 NO10 exclude them from the analysis D 4 Estimated elasticities Table D 3 shows the estimated compensated price elasticities and table C 4 show the income elasticities The own price elasticities confirm that the model satisfies negativity Table D 3 Hicksian Compensated price elasticities Good Mm a O WW G6 m 8 9 aY aA 42 1 Food on which no IVA is levied 0 229 0 039 0 040 0 051 0 087 0 002 0 013
108. ween goods We can then compare the averages of these CVs to averages of the welfare costs of indirect tax changes under the assumption of no behavioral response for broad groups of people e g expenditure decile groups We use a demand model that has been estimated using unadjusted ENIGH data and this simulation has been conducted using the baseline unadjusted data for this reason It is important that the estimation and simulation use the same data e g the same corrections for the under reporting of income or expenditure otherwise the predictions of the demand system will not correspond well with the observed patterns of expenditure in the data used in the MEXTAX simulator Table D 1 lists and describes the 12 goods categories included in the demand system providing detail of the ENIGH expenditure categories included the Bank of Mexico Prices index codes and the assumptions made when prices are unavailable for certain goods Table D 2 lists and describes the demographic variables included in the demand systems 64 Table D 1 Definition of categories used in the demand system Categories Food and drink on which no IVA is levied Food and drink and meals out on which IVA is levied Alcoholic Drinks and Tobacco IVA and IEPS levied Clothing and footwear IVA levied Variable name expnum1 expnum2 expnum3 expnum4 ENIGH codes A001 A197 A203 A214 A215 A216 A217 A218 A242 A198 A202 A219 A222 A243 A247
109. xpenditure distribution losses equivalent to 0 41 for the richest 10 and 0 22 for the poorest 10 on average but slightly regressive when expressed as a fraction of income over the income distribution losses equivalent to 0 26 for the richest 10 and 0 33 for the poorest 10 on average What figures give a more meaningful picture of the distributional impact of the reforms It turns out that the issue of how to group households into decile groups is a separate issue from whether to divide gains losses by income or expenditure A case can be made to group households into decile groups based on their income or their expenditure both are volatile and it is not clear what better captures households underlying purchasing power but as explained in box 5 1 it is most meaningful to divide gains losses from changes to direct taxes ISR and Social Security taxes by zucome and changes in indirect taxes IVA and IEPS by expenditure When examining reforms that affect both direct and indirect taxes such as system 20100 one should consider the impact both as a percentage of income and a percentage of expenditure giving more weight to the former if direct tax changes are most significant and the latter if indirect tax changes are most significant However sometimes it will necessarily be difficult to determine whether a reform is progressive or regressive Box 5 1 Income or expenditure to assess the distribut
110. y et al 2010 2011 and 2012 LATAX is based on IFS micro simulation model for Mexico MEXTAX LATAX draws both upon IFS experts experience in analyzing tax and benefit reforms in the UK and on the experience and methods used by other researchers in Mexico and other Latin American countries MEXTAX is now included built in in LATAX This user manual explains how to use LATAX for analysis in Mexico and includes 1 Introduction what is LATAX 2 Overview of program structure 3 Installation requirements and procedures 4 Running the model a worked example 5 Interpreting outputs of model distribution revenue behavior etc 6 Constructing the input data 7 Appendices which provide details on the reforms modelled appendix A full details on the user editable interface commands and parameters appendix B information on the variables in the input and output datasets appendix C and information on how we estimated a consumer demand system for including in LATAX Most of this manual with the exception of explanations of the consumer demand module and the creation of the input datasets is applicable to use of the LATAX model for El Salvador 1 It is written using a worked example the 2010 Mexican reforms with screen shots to help make it more concrete and to provide a practical example to help people with 1 Introduction The LATAX model is a flexible static microsimulation model developed in

full version - Institute for Fiscal Studies

Contents

Download Pdf Manuals

Related Search

Related Contents