Social Policy Simulation Database and Model (SPSD/M) product overview

Product Description

The Social Policy Simulation Database and Model (SPSD/M) is a tool designed to assist those interested in analyzing the financial interactions of governments and individuals in Canada. It can help one to assess the cost implications or income redistributive effects of changes in the personal taxation and cash transfer system. As the name implies, SPSD/M consists of two integrated parts: a database (SPSD), and a model (SPSM). The SPSD is a non-confidential, statistically representative database of individuals in their family context, with enough information on each individual to compute taxes paid to and cash transfers received from government. The SPSM is a static accounting model which processes each individual and family on the SPSD, calculates taxes and transfers using legislated or proposed programs and algorithms, and reports on the results. A sophisticated software environment gives the user a high degree of control over the inputs and outputs to the model and can allow the user to modify existing programs or test proposals for entirely new programs. The model comes with full documentation including an on-line help facility.

The Database

For the most part, microdata collected by different statistical surveys and various administrative procedures were designed for specific purposes. No one dataset from a given survey or administrative source provides a sufficiently detailed and integrated picture of Canadian households to support the analysis of costs and distributional impacts of the entire tax/transfer system as it moderates the flows of money between governments and individuals. The SPSD is the only such integrated database in Canada.

The SPSD was constructed by combining individual administrative data from personal income tax returns and unemployment claimant histories with survey data on family incomes, employment and expenditure patterns. The techniques used to create the database and avoid confidential data disclosure include various forms of categorical matching and stochastic imputation. While no one record on the database contains information for the same individual from the four bases, the database has been constructed in such a way as to provide a micro-statistically representative sample of Canadians.

Certain adjustments have been made to the data which force agreement between data and known control totals. For example, survey weights have been adjusted to ensure that the population by age and sex corresponds to Census data, and the number of high income Canadians corresponds to the number reported by Canada Revenue Agency. Further adjustments have been made to compensate for item non-response in the surveys (e.g. to increase the number of families receiving Employment Insurance). It should be noted that the SPSD coverage does not extend to the Yukon, the Nunavut or Northwest Territories, persons residing on reservations, or armed forces personnel residing in barracks.

Also included in the SPSD package are effective tax rates, including sales taxes, derived from data from Statistics Canada's Input/Output tables. These rates support the commodity tax model which facilitates the analysis of changes in commodity taxes on Canadian households and individuals.

The Model Algorithms

The SPSD/M comes with the necessary algorithms and parameters to simulate over 20 years of the Canadian tax/transfer system.

The central program, the SPSM, is a micro-simulation based model which calculates taxes and transfers for individuals and families as appropriate. These calculations are performed for everyone on the SPSD and then aggregated to obtain estimates. The SPSM is a static impact model and is therefore not intended to simulate how an individual's behaviour is likely to change in response to various policy options. A second program, the commodity tax model (COMTAX) is a macro-economic Input/Output based model. COMTAX is intended to provide estimates of effective federal and provincial tax rates by province and commodity type. This model is required because many commodity taxes are levied at intermediate stages of production, not at the final retail stage. Rates calculated by the COMTAX model can be input as parameters into the SPSM to obtain estimates of the commodity taxes paid by any given household.

The SPSM is driven by over 2,000 parameters that control three main processes. Control parameters specify input and output files for a specific model run and are used to activate model software facilities. Database adjustment parameters control the inflation or deflation of money items on the database. Tax/transfer parameters control the specific functioning of the tax and transfer programs. Perusing the included lists of the tax/transfer parameters can give the user a feel for the scope of options available in the tax/transfer algorithms.

The SPSM is written and compiled using the C++ programming language. Changes to the tax/transfer system can be made in two distinct ways: the "black-box" or the "glass box" mode. When using the model in the black-box mode, changes to one or more parameters are specified during a user-friendly dialogue. In this mode the user can change levels and rates affecting various benefits, taxes, and eligibility requirements, and can also specify the population that will be modeled. Most applications are developed using this black-box mode. The glass-box mode is designed for use by persons who wish to develop entirely new algorithms or adjust existing algorithms. Glass-box users can write a new, (or modify an existing) algorithm in the C++ programming language and then recompile the complete system using a simple one word command. The user need not be a C++ language expert in order to use the model in glass-box mode, but should be experienced with some high level computer language. The user requires Microsoft C++ to use the model in glass-box mode.

A set of database adjustment algorithms are included with the model. These algorithms allow the user to "grow" the money items on the current database to some future or past year. The user can select a series of growth rates which are then applied by the model. For most variables a set of province specific growth rates are applied. However for certain important or complex variables more complicated rates are applied. For example, employment earnings are grown by rates specific to the province of employment and sex of the individual. These algorithms can be used independently of or in conjunction with alternate yearly demographic weights included with the database.

In the historical time period, the growth factors reproduce, on a provincial basis, the growth of the conceptually closest benchmark series in the System of National Accounts (SNA). In future years, the growth factors reproduce the growth seen in an average of public and private sector forecasts at a Canada level.

The Model Software

The SPSD/M comes with a powerful software support environment designed to enhance the range, speed and ease of policy analysis in both black-box and glass-box modes of use. The comprehensive and flexible set of outputs allow for detailed verification of algorithms and results. The software has been designed to work in an integrated fashion at many different family levels (e.g. household, census family, individual,etc.). A user-friendly interface allows the analyst to view and modify parameters, run simulations and examine results. An on-line help facility provides information on the use of the model dialogue as well as a handy reference to variable and parameter descriptions and lists. Some highlights of the model software are given below.

• A generalized expression facility allows the user to create new variables.
• The user can specify a database subset in terms of any combination of database or modeled variables (e.g. all Ontarians with income below $40,000).
• Marginal tax rate analyses can be performed in a single execution of the model.
• Two distinct tax/transfer systems can be run simultaneously. Base and variant runs may be differentiated on the basis of parameter inputs, alternate algorithms (glass-box), or both.
• Comprehensive set of default tables includes dollar amounts and persons reporting for over 40 variables with breakdowns by province, income class, family type, and proportion above or below specified income levels.
• Built-in custom tabulation routines allow users to specify custom n-dimensional tables with database, model, or user defined variables and statistics. The user is given full control over formats, precision and labeling of the tables.
• Database output interfaces allow the user to create custom files in compressed binary format, ASCII format or as a documented SAS file. These files are for subsequent analysis using the SPSM or other software packages and may contain any database or modeled variables.

Documentation

The documentation for the SPSD/M is necessarily extensive due to the complex nature of the database and model. The majority of the documentation is of a reference nature providing definitions of parameters and variables or descriptions of algorithms and programs. The documentation is written and organized in such a way as to provide the analyst first with a conceptual and functional understanding of the product and its use, and then where to find and how to use the detailed operational specifications included in the manuals. The documentation is consequently organized into fourteen separate guides and grouped into three volumes as follows:

Package Contents and Ordering Details

Full SPSD/M

Cost: As of December 10, 2013, catalogue no. 89F0002XCB has become free. A Licence Agreement is required before the delivery of this product.

The licence includes the current database, software and documentation plus all updates during the product cycle. A new cycle begins with the creation of a new SPSD/M (database) based on the most recent data (every one to two years).