Maintaining and Updating POLIMOD

Document Sample
Maintaining and Updating POLIMOD Powered By Docstoc
					                  Maintaining and Updating
                          POLIMOD
                      Lavinia Mitton and Holly Sutherland

                                                 August 1999




Introduction ................................................................................................................ 2

1. The POLIMOD micro-data base......................................................................... 2
   1.1 Accessing the data ............................................................................................ 2
   1.1.1 MIMAS how to access................................................................................... 3
   1.1.2 CD-ROM how to access ................................................................................ 6
   1.2 The SAS program to derive POLIMOD variables............................................ 6

2. POLIMOD parameter files .................................................................................. 8
   2.1 File types ............................................................................................................ 8
   2.2 P9900.PAR: Policy parameters...................................................................... 10
   2.3 NEWP990.PAR: Additional policy parameters.......................................... 10
   2.4 B978.BAS: Tax base (and other income assessments)................................ 10
   2.5 E990a.PAR: Indirect tax policy parameters ................................................ 11
   2.6 LT956990.PAR: Local tax parameters .......................................................... 11
   2.7 PRI_990.PAR: Retail prices............................................................................ 12
   2.8 ADJ95990.ADJ: Updating from the data year to the policy year............. 12
   2.9 CONTROL.DAT: Global parameters........................................................... 12
   2.10 TAKEUP.DAT: Level of take-up ................................................................ 13

3. Annual and other changes to the POLIMOD program ................................ 13

4. Distributional output.......................................................................................... 13

5. Installing and running POLIMOD .................................................................. 13
Introduction
The purpose of this paper is to describe the process of maintaining and
updating POLIMOD. It is mainly intended as a reference document, internal to
the Microsimulation Unit. However, it may be of wider interest (a) as
background information for POLIMOD users or (b) as a guide to all the
processes, for prospective builders of tax-benefit models.

The intention is to be as general as possible. However, specific details are
provided for 1995/6 data and the policy year 1999/2000. Programs,
documents and spreadsheets mentioned are stored on Lavinia’s PC and
copies are available on network directory N:/DAE/mu-share/maintaining
polimod. Programs which were run via telnet are copied onto the network
directory and on Lavinia’s PC in the directory C:/lavinia/MIMAS.

Chart 1 (overleaf) shows the tasks involved in creating and maintaining the
files used by POLIMOD. The processes of accessing the FES data and deriving
the POLIMOD variables are covered in more detail in Section 1 of this
document. Sections 2 deals with the policy parameter files, updating
parameter files and global parameter files. Section 3 covers maintenance and
updating of the POLIMOD program itself and Section 4 covers POLIMOD output.


1. The POLIMOD micro-data base
The task of creating the input data set for POLIMOD is an occasional one and
carried out depending on resources, but generally undertaken every 4-5 years.

1.1 Accessing the data
The data source for POLIMOD is the Family Expenditure Survey (FES).
Permission to use this data has to be sought from the Data Archive at Essex
University (web address: http://dawww.essex.ac.uk/).

The original data source, the FES variables, are accessed either
a) from MIMAS via telnet, or
b) from a CD-ROM locally.

The POLIMOD variables are next created using SAS, either
a) through MIMAS via telnet, or
b) using SAS for Windows on a local PC,
depending on how the data are accessed. Instructions on both methods
follow.




                                                                               2
                             Access
                            FES data
                                                      Chart 1. Overview of
                                                      inputs to POLIMOD
      Other
     sources
     of data
                             Variable
                            derivation
                                                 Policy
                                               parameters

                                           before        after


                        POLIMOD
                      micro-data base
                                                                      Global
                                                                    parameters



                Updating
               parameters




                                         Tax-benefit calculations




                                              Revenue cost




                                             Distributional
                                                output




1.1.1 MIMAS how to access
One way of accessing the data is to use a service provided by Manchester
University Computing (MCC), called MIMAS. Potential users first need to
register with MCC to obtain a userid and password. Users contact their site
representative for an application form. More information on registration is
available from the MIMAS web site at: http://www.mimas.ac.uk/.



                                                                                 3
To use the service, use telnet to connect to irwell.mcc.ac.uk. Irwell is the name
of the machine at Manchester. It is a UNIX system. Having logged in you type
bash at the $ prompt. This activates the ‘bash shell’ which enables you to use
the backspace key and to use the arrow keys to recall previously submitted
commands. MCC supplies and introductory guide and a card which lists
useful UNIX commands.

     Directory Structure
     This diagram shows the directory structure on irwell. emlm228 and
     emhs117 are individual users. emmicro is a filespace accessible by all
     members of the Microsimulation Unit so that data can be shared.
                                      home


                     emlm228   emhs117    emmicro   other
                                                    users ...



                         (sub-directories ...)


     A user will only be able access data directly from the FES dataset from
     within their own user filespace. However, once variables have been
     extracted to a SAS dataset that dataset can be copied across to emmicro to
     save space.

Accessing the year of data and the particular variables you need is done by
issuing commands and writing and submitting a SIR program. MCC runs
courses on SIR and makes available notes on how to use it. In ‘interactive
mode’, SIR has a menu-driven system which allows you to explore what
variables are available. To access the data itself SIR ‘batch mode’ is used.

The 3 SIR programs which were used to access the 1995/6 variables are:
~emlm228/FES9596/DATA/fes9596d.sir (derived variables)
~emlm228/FES9596/DATA/fes9596r.sir (raw variables)
~emlm228/FES9596/EXPEND/exp9596.sir (expenditure variables)

Open and edit the program with the ‘joe’ editor. MCC supplies a card with
information on how to use the ‘joe’ editor. Having specified which variables
you want, the program is submitted with the command:
bash$         fes filename      /db=fes9596 /ba
Note that the position of the spaces before the backslash is important.

For FES 1995/6 data, there are 2 separate databases – one for the raw
variables and one for the derived variables. The command to access the raw
variables is:
bash$         fes filename      /db=fes9596r /ba
However, for some data years there is one database only of both raw and
derived variables.

It is possible that the first time a user tries to run a SIR program, an error
message will appear because the terminal is not emulating a vt100. If this
happens type the following:


                                                                                  4
bash$          export TERM=vt100

The log of the SIR extraction is written to a file called stdout. This is viewed
by typing:
bash$          joe stdout

The SIR extraction creates a file called data.dat, for example. This is
converted to a SAS dataset using the following commands
bash$          sir2sas data.sas
bash$          sas data

These last two steps need to be repeated for every data file created by the SIR
program. A SAS dataset with the name data.ssd01 will be created. At this
point data.dat, data.sas, data.sas~ and data.log may be deleted to save
space. Further information can be found in the notes on SIR provided by
MCC.

     File Management
     When a user is registered on irwell, their userid is assigned to a group for file
     mangement purposes. Members of the Microsimulation Unit are members of
     the emmicro group.

     File access permissions on irwell can be changed at three levels of access,
     read or copy (r), write or edit (w) and execute (x), and for three classes of
     user , the owner (u), the group (g) and all others (o).

     When you get a listing of files using ls –l, file permissions are shown in
     the following way:
     total 480
     drwxr-xr-x    2   emlm228   emmicro      8192   Jul   21 12:13 DATA
     drwxr-xr-x    2   emlm228   emmicro      8192   Jun   17 16:28 EXPEND
     -rw-------    1   emlm228   emmicro     77868   Mar   30 1998 inc956.log
     -rw-------    1   emlm228   emmicro      5713   Mar   30 1998 inc956.lst
     -rwxr-----    1   emlm228   emmicro     51501   Apr   15 1998 inc956.sas

     The ‘d’ in the first column indicates that it is a directory; the following nine
     characters indicate accesss permission for the file (read, write, execute) and
     are show in the order owner, group, other. Hyphens denote lack of
     permission. emlm228 denotes the owner, and emmicro the group under
     which the files were created.

     To give permission for the group to access a file, the syntax is:
     bash$ chmod g+w (or +r,+x) filename

     To change the group assignment of a file, the syntax is:
     bash$ chgrp groupid filename

     When sharing data or program files among users it is important that file
     access privileges have been set up correctly. Otherwise, programs will not
     run. Shared files need to be assigned to emmicro group and the group
     needs to have read, write or execute permission, as appropriate. Directories
     and sub-directories containing the shared files also need correct group
     assignment and permissions.




                                                                                         5
The SAS datasets created for use in the derivation were
~emmicro/FES9596/DATA/hou_d.ssd01 (household-level derived variables)
~emmicro/FES9596/DATA/hou_r.ssd01 (household-level raw variables)
~emmicro/FES9596/DATA/per_d.ssd01 (person-level derived variables)
~emmicro/FES9596/DATA/per_r.ssd01 (person-level raw variables)
~emmicro/FES9596/EXPEND/exp956h.ssd01 (household-level expenditure
variables)
~emmicro/FES9596/EXPEND/exp956p.ssd01 (person-level expenditure
variables)

Note that the SAS datasets were copied from ~emlm228/FES9596/DATA to
~emmicro/FES9596/DATA, which is the Microsimulation Unit shared space.

1.1.2 CD-ROM how to access
As an alternative to accessing the data via MIMAS, the Data Archive may
supply the variables on a CD-ROM. If the data has been requested in SAS
format, SAS transport files would usually be supplied by the Data Archive.
The instructions provided by the Archive should be followed to convert SAS
transport files to SAS datasets and save them to a local hard drive using the
SAS 'export engine’.

As a general point, note that SAS datasets (suffix .ssd01, .sd2) are corrupted
when copied across machines, so they need converting to SAS transport files
(suffix .xpt) for copying.

POLIMOD   variables would then be derived using SAS for Windows.

     Downloading data from MIMAS
     Sometimes a user may want to bring down a SAS dataset from MIMAS to
     use locally on a PC with SAS for Windows. To download a SAS dataset, it
     must first be coverted to a SAS transport file. An example of a program to
     convert a dataset is:
     ~emmicro/FES9596/EURMOD/eurxprt.sas
     The SAS transport file can be FTP’d in binary mode and is coverted back to a
     SAS dataset in SAS for Windows using another similar program.

1.2 The SAS program to derive POLIMOD variables
The main body of the SAS program to derive the POLIMOD variables will be
the same whether the derivation is done on MIMAS or locally. The main
difference between the two programs would be referencing of the SAS
datasets and external files. MCC run courses and have available guides on
how to use both SAS on MIMAS and SAS for Windows.

For the FES 1995/6 data MIMAS was used. The relevant programs are
~emmicro/FES9596/inc956a.sas (income and characteristics variables)
~emmicro/FES9596/EXPEND/exp956.sas (expenditure variables)

The derivation was done within the emmicro filespace. This is because the SAS
datasets had been moved into emmicro.



                                                                                    6
The transformations of the FES variables to create POLIMOD input are
documented in Microsimulation Unit Research Notes MU/RN/28 (income
and characteristics variables) and MU/RN/29 (expenditure variables).

     In addition to FES variables, three files of non-FES data were used within the
     programs: self-employment income index, mortgage rates and weights.

     • Self-employment income index, used to derive variable ‘seinc’.
     ~emmicro/FES9596/DATA/seindex.txt
     This file is used to update self-employment income from the last year of
     accounts available to 1995/6. It was created from Excel spreadsheet
     C:/lavinia/polimod/self-emp/seindex.xls
     The data sources used are documented within the spreadsheet.

     • Mortgage rates, used to derive variable ‘grossmi’.
     ~emmicro/FES9596/DATA/mortrat.txt
     This text file was created from Excel spreadsheet
     C:/lavinia/polimod/mortgage/mortrat.xls
     The data sources used are documented within the spreadsheet.

     • Weights, used to derive variable ‘fuwt’.
     ~emmicro/FES9596/DATA/f9596pop.txt
     This file contains family weights for grossing up the FES data sample to be
     represenatative of the whole population. Weights are created using ONS
     population totals and the FES count. The text file was created from Excel
     spreadsheet
     C:/lavinia/polimod/grossing/grossing.xls
     The data sources used are documented within the spreadsheet.

The POLIMOD variables were output as 4 ASCII files, one each at the person,
family and household level, and a household-level expenditure data file. The
ASCII files were output to:
~emhs117/fes9596/per956b.dat
~emhs117/fes9596/fam956b.dat
~emhs117/fes9596/hou956b.dat
~emhs117/fes9596/exp956.dat
These were then FTP’d to a local machine for input into POLIMOD. Summary
statistics on the income and characteristics variables are documented in
Microsimulation Unit Research Note MU/RN/28.

Note that corrections or revisions of the data files lead to the last character of
the file name changing to the next letter in the alphabet (e.g. the next revision
of the person-level data would be named per956c.dat).

The data file names are hard-wired into the POLIMOD program C-code so that
any particular version of POLIMOD can only access a particular version of each
year’s micro database.

A further data file (hwt956.dat) is created locally using ‘fuwt’, the family unit
grossing-up weight. This is converted into a household-level weight using
Joanna Gomulka’s software.




                                                                                      7
  Grossing-up weights
  The local file is hwt956.dat and is created from input file fuwt956.dat
  (which can either be created on MIMAS at the same time as the main
  data files or extracted from fam956b.dat after it has been downloaded).
  This file contains a serial number, household id, id of family unit and
  fuwt. Also required are the population totals used in the SAS
  construction of ‘fuwt’.

  For 1995/6 Joanna’s DOS version of the program was used (she has a
  superior, more flexible Windows version, but the DOS version is set up
  to do exactly what is needed to produce household weights from family
  unit weights). This is in C:/grossing on Holly’s machine. See note.wp in
  that directory for instructions. (This directory is copied into N:/dae/mu-
  share/maintaining polimod/grossing)


The datafiles are stored in a directory C:\polimod\data956. This location is
hardwired into the program C-code, so needs to be the same for any
installation of POLIMOD.

2. POLIMOD parameter files

2.1 File types
There are three types of parameter files:
1. Policy parameter files which must be updated each time policy changes (and
   which may need to be added-to if policy changes or the policy options in
   POLIMOD are extended). When POLIMOD is used there are two copies of
   each type of file, one for before the policy change and one after.
2. Database updating parameter files. These must be changed for each new
   database AND for each new policy year. There are two copies of each type
   of file when POLIMOD is run (so that alternative updating assumptions may
   be compared).
3. Global parameters. These govern the assumptions and operations when
   running the model. There is one set of these (assumptions are the same
   ‘before’ and ‘after’). These files only need ‘maintenance’ if new run options
   are introduced.

Default files of type 1 and 2 are stored in a location that is specific to the
policy year. Type 2 files are named so that the data year to which they apply
is clear. For the 99/00 policy year the location is C:\polimod\par9900. This
location is NOT hardwired into the model code. Instead, copies of the files are
made into the directory from which POLIMOD is being run. TWO copies are
made (with the exception of the price file) so that one (or both) can be edited
to specify a policy. The naming conventions are shown in the tables that
follow. These names ARE hardwired in.




                                                                               8
Copying of default parameter files can either be done individually and
manually, or alternatively, batch files are set up to copy sets of files to be used
together.

Default files of type 3 are present in the directory from which POLIMOD is run.
They are edited as required.

The following 3 tables show the files by type, giving their original names and
those used by POLIMOD to refer to them.

Type 1: policy parameter files
Type name (in 99/00)   POLIMOD name(s)       Function
P9900.PAR              POL_DEF1.PAR          Policy parameters relating to child benefit,
                       POL_CH1.PAR           means tested benefits, basic income
                                             schemes, National Insurance Contributions
                                             and income tax
NEWP990A.PAR           NEWPOL_D.PAR          Policy parameters relating to benefit and
                       NEWPOL_C.PAR          direct tax reforms not yet actually
                                             implemented (but including the
                                             minimum wage)
B978.BAS               BASE_DEF.BAS
                                             Parameters for defining assessable income
                       BASA.BAS
                                             for means tested benefits, National
                                             Insurance Contributions and income tax
E990A.PAR              EXP_DEF1.PAR
                                             Parameters for rates of VAT and excise
                       EXP_CH1.PAR
                                             duties used in calculating taxes on
                                             expenditure
LT956990.PAR           LT956_DE.PAR
                                             Parameters for Council Tax using reported
                       LT956_CH.PAR
                                             amounts of Council Tax
LT945990.PAR           LT945_DE.PAR
                                             Parameters for Council Tax using Council
                       LT945_CH.PAR
                                             Tax Band
PRI_990.PAR            PRICE.PAR             Average retail prices used in calculating
                                             excise duties

Type 2: updating parameter files
Type name (in 99/00)   POLIMOD name(s)       Function
ADJ95990.ADJ           ADJ956_D.ADJ          Interest rates (for imputation of mortgage
                       ADJ956_C.ADJ          capital and amount of savings), and
                                             uprating factors for earnings, other income,
                                             non-means tested benefits and expenditure,
                                             to adjust from FES year (1995/6) to policy
                                             year.

Type 3: global parameter files
Type name              POLIMOD name(s)       Function
CONTROL.DAT            CONTROL.DAT           Parameters for controlling analysis and
                                             output
TAKEUP.DAT             TAKEUP.DAT            Parameters for controlling caseload level of
                                             take-up of income support, family credit,
                                             council tax benefit and housing benefit




                                                                                         9
Producing the parameter files involves doing the necessary calculations in a
spreadsheet, then making a text file which contains the relevant parameters
based on the spreadsheet. The processes by which the default parameter files
for April 1999 policies were created are set out below. Two features are in
common across the types of parameter file.

1. Spreadsheets are used to create the parameters that are required (with the
   exception of NEWPxx.PAR, TAKEUP.DAT, CONTROL.DAT and the .BAS files) and
   text files are written using the Excel “save as” facility.
2. The precise format of the text file is significant, and in particular whether
   variables are integers or floats or character strings needs to be checked
   against previous versions of the same file type.

2.2 P9900.PAR: Policy parameters
This file contains policy parameters relating to child benefit, means tested
benefits, basic income schemes, National Insurance Contributions and income
tax. It is revised every year, usually before the Budget and again after the
Budget in response to any new rates announced.

The workbook used was
C:/lavinia/polimod/parameters/P_file.xls
The front sheet ‘main’ has a complete year-to-year record of policy
parameters. The information listed here can be used to estimate the next
years’ parameters before the Budget. The data sources used are Tolley’s, CPAG
Welfare Benefits Handbook, press releases from the DSS and Inland Revenue
and the CTB/HB circulars.

Different sheets are used to create different text files which are used for
specific pieces of work. For example POL_990A.PAR is the actual parameters for
April 1999 and POL_990O.PAR is the parameters for October 1999.

Before any of these text files is used in POLIMOD, it needs to be copied twice
across to the POLIMOD directory and renamed POL_DEF1.PAR and POL_CH1.PAR,
as explained above.

2.3 NEWP990.PAR: Additional policy parameters
This file contains default parameters that are new or experimental, such as those
for the minimum wage or basic income. It is revised from time to time, in
response to any changes that are announced.

2.4 B978.BAS: Tax base (and other income assessments)
This file contains parameters for calculating assessable income for means tested
benefits, National Insurance Contributions and income tax. It is revised from
time to time, in response to any changes announced. (The 978 in the name of
the current file indicates that it was last changed in 1997/8.)




                                                                               10
2.5 E990a.PAR: Indirect tax policy parameters
This file contains parameters for rates of VAT and excise duties used in
calculating taxes on expenditure paid by households. It is revised every year,
usually before the Budget and again after the Budget in response to any new
rates announced.

The workbook used was
C:/lavinia/polimod/indirect tax/indirect.xls
In the workbook there is a sheet ‘Budget’ that can calculate the revised duties
per unit, when they are reported as a percentage or pence rise. Sheet ‘Vat &
Excise’ gives a record of past rates of indirect tax. Sheet ‘election’ was used to
predict future rates of duty. The data sources used are the HM Customs and
Excise Annual Report and press releases.

Before any of these text files is used as the default policy setting in POLIMOD, it
needs to be twice copied across to the POLIMOD directory and renamed
EXP_DEF1.PAR and EXP_CH1.PAR, as explained above.

Some of the issues that arise in updating the indirect tax files when changes
are announced are noted in C:/lavinia/polimod/indirect tax/issues.doc

2.6 LT956990.PAR: Local tax parameters
The POLIMOD user has a choice to either use the reported amounts of council
tax paid in the data or to simulate council tax from information on council tax
band in the data. The parameter files for both methods are therefore needed.

The workbook used to make the text file for updating amounts of council tax in
the data is:
C:/lavinia/polimod/localtx/956data/9596data.xls
The front sheet ‘coun.tax_all’ is a year-to-year record of average Band D
equivalent council tax. The data sources are documented on the spreadsheet.
The sheet ‘956_990’ calculates an updating factor by which the council tax has
risen between 1995/6 and 1999/0 by region.

Before this text file is used as the default policy setting in POLIMOD, it needs to
be copied twice across to the POLIMOD directory and renamed LT956_DE.PAR
and LT956_CH.PAR as explained above.

The workbook used to make the text file for simulating from council tax band is:
C:/lavinia/polimod/localtx/945data/9495data.xls
The front sheet ‘coun.tax_all’ is a year-to-year record of average Band D
equivalent council tax. The data sources are documented on the spreadsheet.
The sheet ‘coun.tax990’ contains the actual Band D council tax amounts for
1999/0, which are copied to text file LT945990.PAR. This file is copied twice
into the POLIMOD directory and re-named LT945_D.PAR and LT945_C.PAR. (The
parameter file name refers to the data year. However, the file of Band D
council taxes in 1999/0 is the same regardless of the data year so the 1994/5
data year files are also good for use with the 199/6 data year).


                                                                                 11
2.7 PRI_990.PAR: Retail prices
This file contains average retail prices because retail prices per unit of goods
with flat rate taxes are needed to calculate excise duties on expenditure paid by
households.. It is revised every year before the Budget.

The workbook used was
C:/lavinia/polimod/indirect tax/indirect.xls
Sheet ‘prices990’ records the prices which were used.
The data source was average prices calculated on the basis of the prices
charged at ‘Sainsbury’s’. A more detailed description of the method is given
in Microsimulation Unit Research Note MU/RN/29, page 41.

Sheet ‘pri_990’ was used to create the text file PRICE.PAR.

2.8 ADJ95990.ADJ: Updating from the data year to the policy year
Income and expenditure variables in the database need to be updated from
the data year to the policy year. This is usually done every year before the
Budget.

The workbook used to update the FES1995/6 to 1999/2000 was:
C:\lavinia\polimod\updates\Adj9599.xls
The front sheet ‘update’ calculates the updating factors for income variables.
The second sheet ‘expend’ calculates the updating factors for expenditure
variables. The data sources and method used for the updating are
documented in the workbook itself and in Microsimulation Unit Research
Note MU/RN/30. The .ADJ files also contain parameters for simulating the
amount of SERPS in the FES state pension variable. These refer to the data
year and do not need annual updating. Also in these files are under-reporting
factors for self-employment and investment income. These do not need
updating (unless new information becomes available) but the user can change
them as they wish.

This work book was used to create text file ADJ95990.ADJ. The income and
expenditure updating factors are both in this single .ADJ file, and it was
created using the ‘save as’ command in Excel.

Before this text file is used in POLIMOD, it needs to be copied twice across to
the POLIMOD directory and renamed ADJ956_D.ADJ and ADJ956_C.ADJ, as
explained above.

2.9 CONTROL.DAT: Global parameters
This file contains global parameters for controlling analysis and output. The
POLIMOD user makes changes to the values of parameters depending on the
output and assumptions they want in the model run. Maintenance is of
CONTROL.DAT is only needed if new options are to be introduced.




                                                                                  12
2.10 TAKEUP.DAT: Level of take-up
This file contains parameters for controlling caseload level of take-up of income
support, family credit, council tax benefit and housing benefit. TAKEUP.DAT is
revised if there are new DSS take-up estimates. It can be revised by the POLIMOD
user, according the assumptions they wish to make. The current default
parameters are taken from Income Related Benefits: Estimates of Take-up in 1995/6
(DSS: ASD, 1997).

3. Annual and other changes to the POLIMOD program
1. Policy parameter files and updating parameter files are revised each year.
2. Changes in current or prospective policy may require changes in policy
    parameters and/or C-coded policy rules. New policy options will usually
    require changes in policy parameters and/or C-coded policy rules. These
    need to be added sparingly and carefully, taking account of :
   • the read routines and the header files (which contain definitions)
   • old parameter files that may need to be read by the new program (they
       need to conform to a fixed format).
3. Each time a new database is introduced the C-code needs to be amended:
   • to take account of any necessary change in definition or format of the
       variables (changes are minimised);
   • to hardwire the names of the new data files;
   • to offer the user choices over which database(s) to use.
   • If some policy simulations can only be carried out using some data
       years (databases), then this needs to be accounted for in the C-code.
   • In addition, new .ADJ files need to be created, one for each policy year
       that is to be simulated for the new database.
4. The C-coded policy rules get revised as necessary (following real changes
    or discovery of errors) and when new options are needed in the model.
5. The most up-to-date compiled version of the C-program is in the
    appropriately date-named directory on N:/DAE/mu-share. The
    corresponding source is in a sub-directory.

4. Distributional output
POLITAXM.EXE is the C-coded program that produces standard distributional
output. It is stored in the same directory as the calculator model. It does not
need regular revision.

5. Installing and running POLIMOD
Instructions for installing and running POLIMOD are provided separately.
Prospective users need to have secured from the Data Archive permission to
use FES data for a specific purpose.

See Microsimulation Unit Research Note MU/RN/19 for an outline of the
options and output that are available.




                                                                                  13
                   MICROSIMULATION UNIT PUBLICATIONS

                               Discussion Papers

MU9201   Stephen Pudney: Dynamic Simulation of Pensioners' Incomes: Methodological Issues and
         a Design for a Computer Model for Great Britain
MU9202   Stephen Pudney and Holly Sutherland: The Statistical Reliability of Microsimulation
         Estimates: Results for a UK Tax-Benefit Model
MU9301   A B Atkinson, Karen Gardiner, Valérie Lechene and Holly Sutherland: Comparing Low
         Incomes in France and the United Kingdom: Evidence from Household Expenditure
         Surveys
MU9302   Carlos Farinha Rodrigues: Measurement and Decomposition of Inequality in Portugal
         1980/81- 1989/90
MU9303   Andrea Brandolini: A Description and an Assessment of the Sample Surveys on the
         Personal Distribution of Incomes in Italy
MU9401   Sheila Marnie and John Micklewright: Poverty in pre-reform Uzbekistan: What Do Official
         Data Really Reveal?
MU9402   Stephen Pudney and Holly Sutherland: Statistical Reliability and Microsimulation: The Role
         of Sampling, Simulation and Estimation Error
MU9403   Jean-Yves Duclos and Magda Mercader: Household Composition and Classes of
         Equivalence Scales: With Application to Spain and the UK
MU9501   Holly Sutherland: The Development of Tax-benefit Models: A View from the UK
MU9502   Gordon Keenay: Personal Income Tax Modelling in the UK Inland Revenue
MU9503   Holly Sutherland: Static Microsimulation Models in Europe: A Survey
MU9504   Jφrgen Aasness, Erik Fjærli, Hanne Gravningsmyhr, Ann Marit Kleive Holmφy, Bård Lian:
         The Norwegian Microsimulation Model LOTTE: Applications to Personal and Corporate
         Taxes and Social Security Benefits
MU9505   Moira Wilson: Earnings Distributions from the Family Expenditure Survey and the New
         Earnings Survey compared
MU9506   A B Atkinson, L Rainwater, T Smeeding: Income Distribution in European Countries
MU9507   Vani K Borooah and Grainne Collins: Was there a Regional Dimension to Changes in
         Income Inequality in the UK over 1982-1992? An Analysis Based on a Joint Decomposition
         of Income Inequality by Region and by Employment Status
MU9508   Gerry Redmond and Holly Sutherland: How has Tax and Social Security Policy Changed
         since 1978? A Distributional Analysis
MU9601   Holly Sutherland: Households, Individuals and the Redistribution of Income
MU9701   A B Atkinson in conjunction with F Bourguignon, T Callan, L Chauvel, F Cowell, J Hills, C
         Joyeux, D Meulders, B Nolan, R Plasman, G Redmond, N Rossi and H Sutherland:
         Measurement of Trends in Poverty and the Income Distribution
MU9702   N A Klevmarken: Modelling Behavioural Response in EUROMOD.
MU9703   F Bourguignon, C O’Donoghue, J Sastre-Descals, A Spadaro and F Utili: Eur3: A
         Prototype European Tax-Benefit Model
MU9704   R Hancock: Computing Strategy for a European Tax-Benefit Model
MU9705   Holly Sutherland (ed): The EUROMOD Preparatory Study: a Summary Report.
MU9801   A B Atkinson and H Sutherland: Microsimulation and Policy Debate: A Case Study of the
         Minimum Pension Guarantee in Britain
MU9802   Francesca Gastaldi and Paolo Liberati: Towards a two-rate VAT in Italy: Distributional and
         Welfare Effects




                                                                                      14
MU9803     Cathal O’Donoghue and Martin Evans: Recasting Safety Nets: Reforming Social
           Assistance in Germany, Ireland and the United Kingdom
MU9804     Holly Sutherland: A Citizen's Pension


                                   Research Notes

MU/RN/1    Holly Sutherland: The Immediate Impact of a Minimum Wage on Family Incomes, August
           1991
MU/RN/2    Holly Sutherland and Gerry Redmond: The Impact of the 1993 Budget Plans on the
           Distribution of Household Incomes, March 1993
MU/RN/3    Gerry Redmond and Holly Sutherland: How to Raise Two Billion: Some Alternatives to VAT
           on Domestic Fuel, September 1993
MU/RN/4    Joanna Gomulka: Grossing up: a note on calculating household weights from family
           composition totals, March 1994
MU/RN/5    Gerry Redmond, Holly Sutherland and Moira Wilson: POLIMOD: An Outline, January 1995
MU/RN/6    Moira Wilson: Simulating Income Support Entitlement from the 1991 Family Expenditure
           Survey, March 1995
MU/RN/7    Moira Wilson: Non Means-Tested Social Security Benefits and the Family Expenditure
           Survey: Disaggregating the 1991 Data, March 1995
MU/RN/8    Holly Sutherland: The Commission on Social Justice's Proposals for Child Benefit: A
           Comment, October 1994
MU/RN/9    Gerry Redmond: POLIMOD: The Derivation of Variables, March 1995
MU/RN/10   Gerry Redmond, Holly Sutherland and Moira Wilson: Taxing Budgets: The Impact of the
           March 1993, November 1993 and November 1994 Budgets on the Distribution of Household
           Incomes in 1995/6, December 1994
MU/RN/11   Gerry Redmond: POLIMOD: The Calculation of VAT and excise duties on household
           expenditure, March 1995
MU/RN/12   Gerry Redmond: Simulating Local Taxation in POLIMOD, December 1996
MU/RN/13   Gerry Redmond: Modelling Employee Class 1 National Insurance Contribution Regimes,
           December 1995
MU/RN/14   Gerry Redmond and Moira Wilson: Validating POLIMOD Output, December 1995
MU/RN/15   Holly Sutherland and Moira Wilson: Using Family Expenditure Survey Data to Simulate
           Income Tax in POLIMOD, June 1996
MU/RN/16   not published
MU/RN/17   Gerry Redmond and Holly Sutherland: The Proposed Earnings Top-up: A Comment,
           September 1995
MU/RN/18   Gerry Redmond and Moira Wilson: Updating POLIMOD, December 1995
MU/RN/19   Gerry Redmond, Holly Sutherland and Moira Wilson: POLIMOD: an outline, Second Edition,
           July 1996
MU/RN/20   Holly Sutherland in conjunction with: AB Atkinson; P Bosi; F Bourguignon; J Le Cacheux; T
           Callan; R Hancock; J Hills; A Klevmarken; D Meulders; N Rossi; A Salomäki; K de Vos:
           EUROMOD: A European Benefit-Tax Model, September 1996
MU/RN/21   Tim Callan and Holly Sutherland: Comparative Analysis of Basic Income Proposals:
           prospects for the use of national tax-benefit models in five European countries, September
           1996
MU/RN/22   Holly Sutherland: Who Benefits from Tax Cuts?, November 1996




                                                                                       15
MU/RN/23   M Chadwick, C O’Donoghue, G Redmond and H Sutherland: Neither Santa Claus nor
           Scrooge: The Impact of the November 1996 Budget on the Distribution of Household
           Incomes, November 1996
MU/RN/24   Holly Sutherland: Policy Simulation at the European Level: A Guide to EUROMOD, April
           1997
MU/RN/25   F Bourguignon, C O'Donoghue, J Sastre-Descals, A Spadaro and F Utili: Technical
           Description of Eur3: A Prototype European Tax-Benefit Model, May 1998
MU/RN/26   Cathal O’Donoghue: Simulating the Irish Tax-Transfer System in Eur6, May 1998
MU/RN/27   Lavinia Mitton: POLIMOD: The Derivation of Variables. 1994/95 FES Edition, Revised
           version August 1999
MU/RN/28   Lavinia Mitton: POLIMOD: The Derivation of Variables. 1995/96 FES Edition, Revised version
           August 1999
MU/RN/29   Lavinia Mitton: POLIMOD: The Calculation of VAT and Excise Duties on Household
           Expenditure. 1994/95 and 1995/6 FES Edition, October 1998
MU/RN/30   Lavinia Mitton: Updating POLIMOD: Second Edition, October 1998
MU/RN/31   Tim Callan, Cathal O'Donoghue, Holly Sutherland, and Moira Wilson: Comparative Analysis
           of Basic Income Proposals: UK and Ireland, February 1999
MU/RN/32   Herwig Immervoll, Lavinia Mitton, Cathal O'Donoghue, Holly Sutherland: Budgeting for
           fairness? The distributional effects of three Labour Budgets, March 1999
MU/RN/33   Lavinia Mitton and Holly Sutherland: Maintaining and Updating POLIMOD, August 1999




                                                                                      16

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:83
posted:2/3/2011
language:English
pages:16