The Treasury

Global Navigation

Personal tools

3  The steady state

The steady-state version of the model comprises 49 exogenous variables. In order to remove the short-term volatility of the data representing these variables, a Hodrick-Prescott (1997) filter (HP) is used to decompose the exogenous variables into trend and cyclical components. The trend variables are then used to solve for the steady state solution over the historical period. The HP filter calculates a trend component of a variable by minimising the following expression:


where yt is the log of the actual series and τt is its trend. The parameter λ is a smoothness constraint that determines how closely trend output follows the actual series.

In Section 3.1, we discuss the historical properties of the key exogenous variables. In Section 3.2 we compare the steady state solution of the model with actual values for two key endogenous variables, namely private sector production and real exchange rates.

3.1  The exogenous variables

3.1.1  NAIRU

Figure 5 shows the actual unemployment rate and its trend value estimated using the HP filter. The historical trend value of the unemployment rate is used as a proxy for the Non-Accelerating Inflation Rate of Unemployment (NAIRU). As can be seen from Figure 5, while the estimated NAIRU rose in the late 1980s and early 1990s, since around 1993 it has fallen continuously to around 5.5% in 2001. This fall in the NAIRU may reflect the impact of labour market reform in the early 1990s. The 1991 Employment Contracts Act (ECA) removed compulsory unionism and promoted wage bargaining on the basis of individual employment contracts

In NZTM’s forecast environment, the steady-state assumption for the NAIRU is set at 5.5%.

3.1.2  Labour Force Participation Rate

Figure 6 shows actual and trend labour force participation rates. The participation rate is defined as the ratio of the labour force to the population aged 15 years and over. In NZTM’s forecast environment, the steady-state assumption for the participation rate is 65.8%. This assumption is based partly on a long-term historical average of around 65% and partly on recent higher rates of participation.

3.1.3  Net Real Household Financial Asset to Real GDP ratio

The steady-state assumption for the net real household financial assets to GDP ratio is a key determinant of the steady-state solution for the net foreign debt to GDP ratio. As can be seen from Figure 7, there is a decreasing trend in the net real household financial assets to GDP ratio between 1987 and 2001. The consequence of the trend decline in this ratio is that the net foreign debt to GDP ratio increased between 1987 and 2001 (see Figure 8), reflecting the fact that the New Zealand economy has run current account deficits over this period. Haugh (2001) has argued that the increase in the current account deficit from 1997-2000 and hence the increase in the net foreign debt to GDP ratio after 1997 was due to households undertaking residential investment. In other words, the household sector as a whole was not saving enough to fund its investment in housing and had to borrow the shortfall. Banks went offshore to meet this shortfall in funding

In NZTM’s forecast environment, the steady-state assumption for the net real household financial assets to GDP ratio is set at –0.9. This assumption leads to a level of current account deficits as a proportion of New Zealand GDP of 4.3%. This is consistent with the average current account ratios and underlying net capital inflows over the 1990s.

3.1.4  Real Public Liability to GDP ratio

The model-based estimate of the real public liability to GDP ratio is shown in Figure 9. While the public liability to GDP ratio rose during the late 1980s and early 1990s, this ratio declined between 1993 and 2001. This trend decline between 1993 and 2001 reflects fiscal consolidation over the 1990s.

The steady-state assumption for the public liability to GDP ratio in the forecast environment is set at 30% to reflect the current public gross debt target.

3.1.5  Export and Import Prices

As shown in Figures 10 and 11, export and import prices in foreign currency did not exhibit a marked long-term trend over the data period chosen. Over the same period, the price of the domestic good has been trending upwards. The difference in productivity between the traded and non-traded sector might explain the divergent price movement between traded and non-traded goods.

The trend growth rates of π1 and π2 are estimated to be negative in the production block, reflecting the increasing openness of the New Zealand economy. In the forecasting horizon,π1 and π2 are set at –0.01 allowing for higher import penetration and more open economy. In order words, the growth rate of the tradable sector is higher than that of the non-tradable sector by 1.0 % per annum. In order to maintain a constant expenditure share of traded goods, the trend growth rates for both export and import prices are set equal to 0.5% per annum if the domestic inflation rate is set at 1.5% per annum

Page top