• Summary
• Downloads
• Explanatory Notes
• Related Information
• Past Releases

To Our Subscribers:

Following the introduction of new National Account benchmarks in the September quarter 2005 quality issues were identified with some of the National Income Forecasting (NIF) series in the Modellers' Database. This has resulted in a wider review of all NIF series. The review will investigate the models, assumptions and input data currently being used to create the NIF series.

Most NIF series have acceptable quality and the ABS will continue to release them regularly. The release of the remaining series will be determined on a case by case basis depending on the outcome of the review.

The following NIF Series will not have any data available for this quarter:

TABLE 2. NIF Demand And Expenditure - Seasonally Adjusted ($ million)
Statistical Discrepancy (cvm)

TABLE 3. NIF Private Gross Fixed Capital Formation - Seasonally Adjusted ($ million)
Plant & Equipment : Corporate
Plant & Equipment : Unincorporated

TABLE 4. NIF Depreciation & Deductions - Original ($ million)
Corporate Trading Enterprises

TABLE 5. NIF Interest Payments - Seasonally Adjusted ($ million)
Government : Total
Unincorporated Trading Enterprises
Corporate Trading Enterprises (Seasonally Adjusted)
Persons : Total excl dwellings (Seasonally Adjusted)

TABLE 7. NIF Stocks Of Physical And Financial Assets - Seasonally Adjusted ($ million)
Total Capital Stock : Corporate
Other Building & Structures : Corporate
Plant & Equipment : Corporate
Corporate Sector Debt Capital (Original)
Total capital stock : Corporate (cvm)
Motor Vehicles (cvm)
Other Buuilding & Structures : Total (cvm)
Household Durables (cvm)
Plant & Equipment : Total (cvm)

TABLE 14. NIF Ratios - Seasonally Adjusted
Incomes : Dividends paid Overseas to YDPC$
Building & Construction to Plant & Equipment by pfe

TABLE 15. NIF Taxes & Transfers - Seasonally Adjusted ($ million)
Net Income Taxes Paid : Persons
Payroll Tax : Total
Sales Tax

TABLE 16. NIF Incomes - Seasonally Adjusted ($ million)
Dwellings Rent Received by Persons
Transfers to Overseas

TABLE 17. NIF Budget Sector - Seasonally Adjusted ($ million)
Company Tax Payments
Total Receipts
Subsidies

TABLE 19. NIF Miscellaneous - Seasonally Adjusted
Wage Share of GDP at Factor Cost (ratio)

We apologise for any inconvenience it may have caused you.

IDENTIFIER	DESCRIPTION	UNITS	DECIMAL	COMMENTS
Additions	See note below
Deletions	All TRYM Experimental Industry Series
Changes	There will be a total of 39 tables instead of 56 in the Modellers' database
Note:

Ever since its initial release in 1993, the TRYM model has been subject to a continuing process of review and modification. Documentation of the model is available on the Treasury website, http://www.treasury.gov.au. The main documentation is in The Macroeconomics of the TRYM Model of the Australian Economy, which gives an overview of the model’s structure, and Documentation of the Treasury Macroeconomic (TRYM) Model of the Australian Economy, which gives equation-by-equation details of the model. These documents can be located on the Treasury website, either by typing the keyword ‘TRYM’ in the site’s search function, or by choosing Business\Economic from the menu on the home page. These two documents describe the model as it stood at the time of the 1996 TRYM conference, except for updates to reflect changes to the private-business-sector and labour-market equations outlined in TRYM Working Paper No. 20. Downes, P. and K. Bernie (1999), ‘The Macroeconomics of Unemployment in the Treasury Macroeconomic (TRYM) Model’, TRYM Related Paper No. 20. The version of the model currently available to the public reflects a number of additional changes. The most important of these are:

• Adjustments for the change in the National Accounts from an SNA-68 to an SNA-93 basis, including the use of chain-volume measures.

• Adjustments to tax formulas to allow for the introduction of GST in 2000.

• Reformulation of the dwelling-sector equations as suggested in Jovanoski, Stoney and Downes (1997). Jovanoski, S., N. Stoney and P. Downes (1997), ‘Modelling the Dwelling Cycle in the Treasury Macroeconomic (TRYM) Model’, Paper presented to the 26th Annual Conference of Economists, Economic Society of Australia, Hobart, 29 September to 1 October. This includes the introduction of a new equation for rental vacancies.

• Changes to equations relating to the external sector of the model as a result of a review prompted by the Asian financial crisis. The changes include introducing variables for the SDR-weighted exchange rate and energy prices, specifying behavioural equations for the world price of commodity exports and the price of non-commodity exports, and introducing short-run dynamics into the exchange-rate equation.

As a result of the latest internal review, further changes are being made to the model and to data construction.

• The private-business sector and government-enterprise sector have been combined into a single ‘enterprise’ sector. The behavioural equations for this sector, borrowed from the previous private-business sector, have been reviewed. Changes include removing capital-augmenting productivity growth from the production function and respecifying the dynamics of the price equation.

• The representation of taxes has been reviewed. GST is no longer treated separately from other taxes on products and a distinction is made between the National Accounts measure of consumption of fixed capital and depreciation allowed as a deduction in the calculation of taxes on capital income.

• The default interest-rate reaction function, previously based on the growth rate of the money supply, has been replaced by a modified Taylor rule.

• The impact of the switch to chain-volume measures has been reviewed. In particular, chain-volume corrections have been introduced into the capital-accumulation equations.

• The relative-price-block equations have been reformulated - they now allow only for the impact of import prices on relative expenditure prices. The deflator for ownership-transfer costs (real-estate transfer expenses) has been dropped from the relative-price block, and the equation for the price of non-commodity exports now parallels the relative-price block equations.

• The construction of price deflators has been reviewed to distinguish more carefully between basic and market prices.

• Several new variables have been introduced for quantities that were previously represented only by formulas within equations.

• Expressions for trend corrections, residuals, interest rates and growth rates have been overhauled to ensure greater consistency across equations.

The most significant of these changes are the ones relating to the business sector, taxes and the interest-rate reaction function. They are discussed in more detail in the following section.

Discussion of Major Changes

The Private Sector Production Function

As part of the review of TRYM, the three private business sector production function-related equations were reviewed and substantial changes were made. The three equations are the labour demand equation, the investment equation and the non-commodity price equation.

The private business sector production function specification in TRYM (as it stood at June 2002) was problematic. The most evident problems, all largely interrelated, included an inconsistency between the level of the model’s steady-state and historical business capital stock measures, and upward trends in the business q-ratio and the capital output ratio (on a balanced-growth path these should be constant). There were also issues related to the treatment of the GST and the relationship between commodity prices and producer prices in general.

A number of these problems were directly related to issues around the measurement of the business capital stock (which has been overhauled) and the institutional framework, in which the government-enterprise sector was separated from the private-business sector. Privatisation of government enterprises has tended to increase the capital-output ratio of the private-business sector. The distinction between these two sectors has become increasingly ill-defined in recent years and data issues have become more troublesome. Moreover, the large changes in ownership have dramatically altered the nature of businesses remaining in the government enterprise sector. It therefore seemed sensible to combine the government-enterprise sector with the private business sector into what is now called the enterprise sector.

In the past the rising capital-output ratio of the private-business sector was explained in the model by negative capital efficiency growth. This was required despite the steady fall in the price of business investment relative to the price of business output. The move to an integrated enterprise sector and improvements in the measurement of the capital stock have addressed this issue.

Taxes

The TRYM review has included an examination of the treatment of taxes.

In previous versions of TRYM, taxes have been accounted for in a fashion that was economically sensible but did not attempt to mirror national accounts principles. The model now includes all the National Accounts tax values in an internally consistent manner. Indirect taxes, in particular, have undergone a complete review and are significantly different from the previous versions of TRYM.

There has also been review of direct taxes. The measures of capital income and capital taxes in the previous version of the model implied implausibly high rates of tax on capital income over history. This has been addressed by reassessing the way the National Accounts measure of gross mixed income is split between labour and capital income, and by the use of a depreciation deduction based on tax data in place of the (higher) National Accounts measure of consumption of fixed capital in determining the capital-income tax base.

Monetary Policy Reaction Function

The TRYM model contains a default monetary policy response mechanism. This is required to close the model and is not intended to represent how monetary policy is or should be conducted. Users of the model are encouraged to substitute their own monetary policy specification if they think that is appropriate.

The existing default monetary policy uses an inverted money-demand function and a money-supply growth rule. Since the mid-1990s, the fit of the model’s money-demand function has steadily deteriorated, and the estimated value of the interest-rate elasticity of money demand has increased. As a result, interest rates respond to shocks with a considerable lag under the default monetary policy. For the model to converge to its steady state, however, it is important that interest rates respond in a timely manner. (This property is shared by many other models.) To correct the problem, the existing default monetary policy reaction function has been replaced by a modified Taylor rule. Under this rule, the monetary authority sets interest rates on the basis of inflation and output targets, as in a standard Taylor rule, and also a price level target. Without the exogenous price-level target, the steady-state level of prices would be indeterminate given the pseudo-rational expectations assumption of the model.

Instead of using the modified Taylor rule, users might prefer to respecify and re-estimate the money demand function, impose a more appropriate interest-rate elasticity on the existing money demand function, use a standard Taylor rule with full rational-expectations, or substitute their own monetary-policy specification.

If you have any concerns regarding the changes, please email me at valentin.valdez@abs.gov.au.


Valentin M. Valdez
Database Manager
Analytical Services Branch