3 Scenarios

The immigration scenarios presented in this paper illustrate the impact on the 2021 New Zealand economy as a result of different levels of immigration inflow, different skill compositions of the inflows and/or alternative assumptions regarding economic behaviour. The experiments can be grouped into four broad categories.

First, there are two scenarios where the size of the inflow is altered, one with an increased immigration inflow, scenario A, and one with a reduced inflow, scenario B (see Table 3.1). The composition of the change in labour in each of these scenarios is determined by the modelled calculations of labour requirements (ie, demand-driven). That is, the composition of the labour supply matches the demand for labour that arises from each scenario. In addition, scenario H is the short-run accompaniment of scenario A showing the input after five years (2011) and is summarised in Appendix D.

Secondly, an increased immigration scenario is simulated where the skill composition of the additional labour supply is directly specified to focus on skilled occupations such as managers, professionals, associate professionals, and technicians and trades workers (scenario C). This specification can be seen as mimicking some change in policy direction or external impetus.

Thirdly, in three further scenarios the increase in immigration in scenario A is accompanied by additional influences. Scenario D incorporates an assumption of increased productivity accompanying an increased immigration inflow. This scenario assumes productivity across the economy is 1 percent above the baseline. This productivity improvement can be described as the same level of output being able to be produced with a 1 percent smaller (than baseline) quantity of labour and capital resource.

Scenario E includes an assumption of increased exposure to external trade opportunities accompanying an increased immigration inflow. The increased exposure can be interpreted as a 2 percent (horizontal) shift in the demand for imports and a 1 percent shift (horizontal) in the demand for exports by consumers only (ie, the demand for capital and intermediate imports was not directly changed by this scenario).Scenario F builds on scenario E with an added assumption of flatter export and import demand curves to mimic growing external exposure. That is a change in price costs will have a larger impact on the amount sold off-shore.

Fourthly, scenario F combines the impacts of an increased immigration inflow comprising specified skill composition accompanied by selective productivity improvements and increased exposure to external trade.

As stated in section 2.5 the impacts are described as changes compared with the baseline level of a range of economic measures for 2021. The baseline, can be interpreted as a business-as-usual scenario and assumes an average annual net immigration inflow of 20,000. Productivity and export market growth in the baseline are assumed to be similar to that experienced over recent years. The baseline projects GDP growth at an average 3.1 percent per annum over 2006 to 2021, with full-time equivalent employment growth of the order of 1.5 percent per annum. Further details of the baseline scenario are in section 2.7.

Table 3.1 Scenario list

Scenario	Description
O	Assumes average annual net immigrant inflow of 20,000 (36,000 in and 16,000 out). Composition (in terms of household mix) is similar to that experienced from 1991 to 2006. Implicit migrant skill mix is totally model determined.
A	Assumes additional 20,000 average annual inflow on top of baseline (56,000 in less 16,000 out equals 40,000 inflow). Closure assumes no change (on baseline) in capital-to-labour ratio, aggregate investment-to-gross domestic product (GDP) ratio, and government consumption demand-to-GDP ratio; no change in relative wage rates (on baseline) means that the migrant skill mix is totally model determined.
B	Assumes zero immigration inflow (0 in less 16,000 out = 16,000 net outflow). Closures as for scenario A.
C	As for scenario A, but additional migrants assumed to have skill mix similar to that in 2001 to 2006. Occupation-relative wage rates are model determined.
D	As for scenario A, but with additional 1 percent positive technical change as proxy for economies of scale and productivity improvements.
E	As for scenario A, but with increased propensities to trade (exports and imports), reflecting more open trade situation.
F	As for scenario A, but with refinements incorporating selective skill mix, accompanied by selective economies of scale and trade improvements (G = A + C + D + F).
SHORT1	As for scenario A, but short-run model closure; that is, no change on baseline in physical capital stocks, so steep marginal costs curves (rates of return are model determined).
AUST2	Simulation on baseline to mirror Australian report3 (ie, equivalent to a 50 percent increase in inflow of skilled migrants).

Notes

1. See Appendix D for results.
2. See Appendix E for results.
3. Productivity Commission (2006).

3.1 Migration assumptions

The levels of immigration and changes to populations imposed in each scenario are summarised in Table 3.2. In the increased immigration scenario A, it is assumed that the net inflow of overseas born rises to an average 40,000 per annum. All other factors constant, this implies an average net PLT inflow of 30,000 per annum.

For the zero inward immigration scenario (B), the ongoing outward flow of overseas born means an average net outflow of overseas born of 16,000 per annum. This translates to an average net PLT outflow of 26,000 per annum. An average natural increase of 25,000 per annum implies a reduction from 2006 to 2021 in the total New Zealand population in this scenario.

Note that it is assumed that the changes in the migration flows take five years to adjust to the new levels. The full change in the flow occurs in 2011. In scenario A the inflow increases by 2,500 per year from 10,000 in 2007 to 20,000 in 2011. For scenario B, the inflow falls by around 10,000 per annum.

Table 3.2 Migration 1991-2006 and assumptions in baseline and scenarios A and B¹

		1991-2006	Baseline	Scenario A	Scenario B
NZ natural increase2	x	30,000	25,000	25,000	25,000
Net migration of NZ-born	y	-20,000	-10,000	-10,000	-10,000
Change in NZ-born in NZ population	x+y	10,000	15,000	15,000	15,000
Net migration of overseas born	z	36,000	20,000	40,000	-16,000
Net permanent and long-term inflow	y+z	16,000	10,000	30,000	-26,000
Total change in NZ population	x+y+z	46,000	35,000	55,000	-1,000

Notes

1. Scenario A would have an impact on the natural increase, because migrants would have children in New Zealand. However, for simplicity, the comparison between scenario A and B assumes this is not the case.
2. Births minus deaths.

3.2 Additional assumptions

The productivity and world market growth assumptions are the same as in the baseline scenario. Other assumptions are consistent with earlier studies^[30] with an unchanged capital-to-labour ratio,^[31] and investment and government consumption demand being fixed relative to GDP. The additional labour resources are assumed to be accompanied in the long run by extra accumulation of physical capital resources. The assumption that the aggregate long-term labour-to-capital ratio remains unchanged ensures the experiment captures the impact of the increased immigration alone. Without such an assumption, the experiment's results would reflect a mixture of impacts of increased immigration and a predetermined shift to labour-intensive activities.

In each of the scenarios government spending on health and education is related to (in real terms) changes in relevant populations. In particular, real education spending per person aged 15 years and under is assumed the same as in the baseline. Similarly, real health spending is related to the number in the population not in the labour force,^[32] including those aged 15 years and under.

In addition, the scenarios assume the demand for owner-occupied dwellings is consistent with the number of households, adjusted by tenure changes, as projected in work for the Centre for Housing Research Aotearoa New Zealand and Department of Labour.^[33] The report from this work found that household status (single/couple) and not birthplace (migrant/New Zealand born) was the major determinant of housing behaviour. The report found that recent migrants were more likely to rent homes than the New Zealand-born population, but that longer term, recent migrants' rent/ownership levels were similar to those of the New Zealand-born population. It also found that the capacity of the building industry appeared to be adequate to meet the level of housing demand to 2016, even under a high immigration scenario, as long as the type of accommodation built changed to meet changed demand; that there would be a growing demand for private rental market dwellings; and that the proportion of people living in flats or apartments was likely to increase.

3.3 Summary of how immigration impacts on the economy

Immigration has impacts on both demand and supply sides of an economy (summarised in Figure 3.1). Thus, it is critical that investigations are undertaken using economy-wide models that capture the interaction of both sides of the economy. The investigation should also capture responses to changes in prices (including wages) prompted by different levels of immigration. In this sense, the CGE model is an ideal tool to analyse the impact of alternative immigration scenarios.

The overall economic impact of immigration comprises a balance between the impacts on the demand side and the supply side. In an increased immigration scenario, the additional demand for goods and services arises from the extra households now resident in the country. This additional demand may be modified according to the household, or family, composition of the additional residents. On the supply side, households supply additional productive resources in the form of labour. This additional labour may also be modified according to different skill types.

The combination of additional demand for goods and services and additional labour resources will require, simultaneously, additional machinery, equipment, buildings, and other productive capital. This further requirement will be reflected in increased demand for investment goods. As noted earlier, our immigration experiments are conducted under the assumption that the ratio of labour-to-physical capital remains unchanged in aggregate.

On the one hand, households and investment demand for goods and services are increased. In particular, sectors associated with the production and supply of physical capital resources (investment goods) will benefit from the increased demand for such resources. On the other hand, the additional labour and capital available will be able to supply more goods and services. The balance between these two impacts will determine changes in prices, and so set off further consequential impacts.

Figure 3.1 Schematic of economic impacts of immigration

Because of the increased resources available to New Zealand producers, the price of New Zealand commodities compared with overseas-made goods and services will decline. Consequently, New Zealand producers competing against overseas products (whether domestically or abroad) will also be advantaged.

It is the outcome of the balance between demand and supply or resources and its impact on sectors (whether export or investment oriented) that drive the overall and detailed results.

The scenarios are undertaken under the assumption that the ratio of labour to physical capital^[34] available does not change in aggregate. That is, the increase in labour supply arising from an increased inward immigration scenario is accompanied by a similar increase, over the long term (year 2021), in capital stock. This ensures the model experiments are not predetermined in favour of labour-intensive industries and sectors. Note, however, that the labour-to-capital ratios for individual sectors may change depending on demand for products and the relative costs of these factors of production.

On the other hand, if the assumption of an unchanged aggregate labour-to-capital ratio is dropped, there would be a predetermined shift to labour. In such a case, the results of a model experiment would capture a mixture of the predetermined factors as well as the impact of increased immigration.

---

Footnotes

^[30] Such as Poot et al (1988).

^[31] The ratio of the value of capital equipment to the total amount of employed labour.

^[32] Bearing in mind the predominant age groups for health expenditure are the very young and the very old, the ‘not in the labour force’ group is used as a proxy for those aged 65 and over.

^[33] Sanderson et al (2008).

^[34] That is, non-labour productive resources such as plant, machinery, other equipment, and buildings.