The Treasury

Global Navigation

Personal tools


A Comparison of the NZTM and FPS Models of the NZ Economy - WP 03/25

2  Overview of the NZTM and FPS models

2.1  Overview

The section provides a brief overview of the main similarities and differences of NZTM (Szeto, 2002) and FPS (Black et al., 1997). According to Pagan (2003), FPS is known as an incomplete dynamic stochastic general equilibrium model (IDSGE) and NZTM is known as a Type II hybrid model[1].

In both types of models, the structure of the economy is described by a solid microeconomic theoretical framework that determines how economic agents make their decisions. The main advantage of these types of models is that they are better at dealing with regime shifts. A change in the regime should not affect a “fundamental” set of parameters that determine the shape of functions being optimised by economic agents. This means that these types of models are less subject to the “Lucas critique” (Lucas 1976) compared to other large traditional macro models.

This theoretical underpinning leads to models which can trace out an equilibrium path along which the economy evolves. The NZTM equilibrium path is constructed explicitly using a steady state version of the model. In reality, it is unlikely that the economy would evolve along this equilibrium path as it is continuously bombarded by shocks. However, both models assume there is a long-run tendency for the economy to converge towards a balanced equilibrium path.

2.1.1  Adjustment

A notable difference between FPS and NZTM is the approach to the determination of the adjustment path when the economy deviates from its equilibrium. For NZTM, the dynamic adjustment structure is based on a partial adjustment process. Partial adjustment occurs when economic agents cannot fully adjust to changes in the short run. The adjustment process is rather mechanically and empirically driven without reference to economic theory. In contrast, the dynamic adjustment structure in FPS is formulated on the basis of adjustment costs. Economic agents are faced with two types of cost: the cost from not being at the “equilibrium” value, and the costs of adjusting to that equilibrium value. The adjustment path is chosen to minimise an implicit cost function (Black et al., 1997).

2.1.2  Production and supply

In both models, the behaviour of the representative firm is based on the assumption that firm’s desire to maximize profits for given costs and prices, subject to an underlying production function. In FPS, the economy produces a single good and the production technology is Cobb-Douglas with labour and capital as the inputs. The single good is differentiated by a system of relative prices.

In NZTM, the supply side of the model has a more comprehensive structure, comprising a three-good economy with two distinct traded goods (exports and imports) and one non-traded good.

Prior to 1970, some models including Swan (1955) and Salter (1959) treated exportables and importables as a composite tradable good and other models use the term of trade to represent the external price. These models limited their capabilities to analyse the impact of external price shocks. During the 1970s, the adverse changes in the terms of trade due to oil price shocks faced by many countries and the increasing importance of commercial policy on the terms of trade led to the development of the three-good model that emphasises the distinction between imports, exports and non-traded goods. Example of such models are Dornbusch (1974), Edwards and van Wijnbergen (1987), and Murphy (1988).

The importance of making the distinction between exports and imports for the New Zealand economy was demonstrated by Wells and Evans (1985) and more recently by Buckle, Kim, Kirkham, McLellan and Sharma (2002). Both studies provide evidence that the effects of import and export price shocks on domestic real output and inflation are different.

Another reason for adopting a three-good model is that a substantial proportion of New Zealand’s exported goods are primary based products, which are not easily transformed into domestic goods in the production process. Furthermore, Buckle, Haugh and Thomson (2001) highlight the significance of the primary sector as a source of volatility for the New Zealand economy. They estimated that the primary sector, at round 10 percent of GDP, has similar and sometimes a higher weighted standard deviation of its growth rate than the manufacturing sector and even the services sector, which accounts for around 50 percent of GDP. Buckle et al. (2002) further suggested that the volatility of primary output could be partly attributed to climatic shocks.

NZTM treats imports as an intermediate input in the production of exports and domestic goods. The majority of imports in fact consist of capital goods and intermediate goods. Also imports, which are ultimately consumed, must go through transporting, wholesaling and retailing before meeting final demand.

In NZTM, the production block consists of a nested constant elasticity of substitution function, which combines three inputs (capital, labour and imports) to produce one unique output. This unique output is then split into domestic goods and exported goods using a constant elasticity of transformation function. This production structure assumes that input-mix and output-mix decisions can be made independently.

2.1.3  Consumption

The FPS model has two types of households: 30% are “rule-of-thumb” and 70% are “forward-looking”. “Forward-looking” households maximize their expected discounted lifetime utility subject to a budget constraint and a fixed probability of death. This type of household is able to plan and achieve a smooth lifetime consumption path. On the other hand, rule-of thumb households are unable to smooth their consumption and spend all their disposable income each period (see Mankiw and Campbell, 1990).

In NZTM, consumption is based on a model of households that choose to consume at a level consistent with their income and wealth in the long-run. The consumption decisions are assumed to approximate the life-cycle hypothesis developed by Ando and Modiglianli (1963). In the short term, the dynamics of consumption are driven mainly by its own lag and the adjustment of consumption towards its long run equilibrium. There are two types of consumption goods: housing services and other consumption. Once households have decided on their total level of consumption, they maximize their utility by choosing the optimal allocation between consumption of housing services and other consumption.

2.1.4  Monetary Policy

Another important agent in the model is the monetary authority. According to the Policy Targets Agreement (2002), the main role of the monetary authority is to maintain a stable general level of prices with an inflation target of 1 to 3 percent on average over the medium term. Both NZTM and FPS have a forward looking monetary policy rule that employs the short-term interest rate as the monetary policy instrument to achieve an inflation target of 2 percent in the long-run. Like the fiscal reaction function discussed below, this is a simple linear characterisation of policy behaviour, rather than an attempt to capture all the complexities of actual policy.

2.1.5  Fiscal Policy

Both FPS and NZTM have a similar fiscal policy response function, which adjusts labour income tax rates to attain a long run debt target. These fiscal reaction functions ensure the government meets an inter-temporal budget constraint. For NZTM, the government is forward looking and targets the gross debt to GDP ratio. The fiscal policy target for FPS is a net debt to GDP ratio while the fiscal policy rule is backward looking.


  • [1]A Type II hybrid model refers to a type of model that contains a well specified microeconomic foundation and a steady state equilibrium path. An IDSGE is similar to the Type II model but the dynamic process is more theoretically based.
Page top