The Treasury

Global Navigation

Personal tools


Monthly Economic Indicators

Special Topic: Measuring New Zealand's net international investment position

Last month's special topic looked at the extent and composition of New Zealand's external debt – a key component of total international liabilities – in a global context. This month, we take a closer look at the way in which the stocks of our international assets and liabilities are measured, and try to quantify some of the known gaps in the reporting framework.

Measurement reliant on surveys...

Like most other statistical agencies, Statistics New Zealand uses a range of quarterly and annual surveys to compile its international investment statistics. These surveys capture the international asset and liability positions of a large proportion of the economy well – particularly the official sector (central and local government and the RBNZ), large and mid-sized businesses including banks, and individual holdings made through managed funds.
There are two areas, however, that are known to be under-covered by the existing survey framework. The first relates to the international asset and liability positions of small and medium-sized businesses. These firms are targeted in the survey framework, but corresponding data on the income they receive from foreign investments suggest that the surveys are not capturing their entire international positions. The second area of under-coverage is the international positions of individuals and retail investors – mainly on the asset side – not otherwise captured by surveys of managed funds. This predominantly includes:

  1. Foreign equity and fixed income securities held either directly or via a nominee/custodian;
  2. Overseas real estate assets;
  3. Individuals' overseas superannuation assets.

...posing some specific problems for NZ

Some of these under-covered areas represent measurement challenges for other statistical agencies elsewhere in the world, particularly the amount of real estate owned overseas by residents, as well as real estate owned by non-residents. However, there are two factors specific to New Zealand that further complicate the task here. The first stems from the small size of New Zealand's capital markets, which means that a comparatively high proportion of retail investment by New Zealanders ends up heading offshore, beyond the scope of the survey framework. The second factor relates to our comparatively high migration flows relative to our population. As a result, measuring the asset and liability positions is more of a challenge here than in other countries.

There have been some attempts to estimate the degree of under-covered assets and liabilities in our international statistics in the past. When Statistics NZ introduced tax data from IRD into the current account income series in 2010, there was some debate as to the implications that these previously uncounted income flows had for the size of the unmeasured stock of investment assets. Comment by Westpac in its review of the 2010 June quarter Balance of Payments release suggested that if the rate of return to the ‘new’ IRD income data was the same as for assets already surveyed, the actual net international investment position (NIIP) could be closer to -60% of GDP, rather than -87% recorded at the time.[1]

We have analysed each of the main areas of under-coverage listed above in turn, focusing on individuals’ assets, to arrive at a ‘bottom-up’ estimate of the degree of under-estimation.

Sizeable amount of under-reported assets...

On the asset side, offshore investments of individuals and trusts are likely to constitute the most significant gap in the measurement of our international assets. Statistics New Zealand introduced backdated estimates of the amount of equity assets held by New Zealanders in Australia into the official statistics in the June 2011 quarter, along with other improvements to the reporting framework too.[2] The combined effect of these changes was to increase New Zealand’s international assets by around 5% of GDP in the 2011 year, which lifted the overall NIIP from -75.1% of GDP to -69.4% of GDP.

Taking the official figure of equity assets held directly in Australia (ie, not via managed funds), along with our estimates of the distribution of New Zealanders’ assets held overseas, we estimate that there is around $10-15bn of directly-held portfolio assets that are not captured in the official international statistics. This covers bond holdings by New Zealanders in Australia, as well as equity and bond positions held in the rest of the world.

Pension assets accrued by New Zealanders who have worked overseas are likely to amount to a significant sum too. Making use of migration data, we estimate such pension assets could be around $2-5bn. Given our strong trans-Tasman links, as well as the existence of compulsory savings in Australia, the bulk of these assets are likely to be concentrated in Australia, followed by the UK.

There are no accurate data to go on with regard to ownership of overseas real estate. However, previous surveys of households conducted throughout the 2000s point to considerable overseas holdings in the region of between $2-4bn. Taking the uncertainty of such surveys into account, and factoring in other expected valuation and exchange rate changes over the period, we estimate that New Zealand ownership of overseas real estate could now amount to around $2-8bn. A summary of our estimates of under-covered international assets is shown in Table 1.

Table 1 – Estimates of under-covered international assets of individuals, trusts and estates
Treasury estimate (NZ$)
Offshore retail portfolios 10 – 15bn
Superannuation assets 2 – 5bn
Real Estate 2 – 8bn
Other (inc. bank deposits and life insurance) < 2bn
Total 15 – 30bn

Source: The Treasury

... offset partly on the liability side...

There are, of course, areas of under-coverage on the liability side that will partly offset some of the known gaps in our knowledge on the asset side. These include any outstanding overseas mortgages owed by New Zealanders which would at least partly outweigh the impact of similarly uncounted overseas real estate assets. Real estate in New Zealand owned by non-residents may in fact be larger than the asset estimate above and perhaps less likely to be offset by mortgages owed to New Zealand banks.

However, given that the bulk of New Zealand’s overseas borrowing is concentrated in the banking and corporate sectors – both of which are captured well by the current survey framework – we are confident that under-reported assets outweigh under-reported liabilities.

...but still points to a ‘true’ NIIP improvement

Overall, then, while our analysis is necessarily subject to a wide degree of uncertainty, we estimate that complete measurement of under-reported international assets and liabilities would result in an improvement in our NIIP of around 5-10% of GDP. This would lift our NIIP from its current level of -72.6% of GDP as at the end of the September 2012 quarter to somewhere in the region of -65% of GDP. Adjusting for the current inclusion of reinsurance claims in the official asset statistics, the ‘true’ NIIP may in fact be similar to the current reported level.

Making further improvements to the data

Making further improvements to the measurement of international investment data over and above those already made by Statistics New Zealand is a difficult task.

Two forthcoming surveys will go some way towards filling some of the gaps discussed here. The first will involve surveying all firms involved in international investment in the March 2013 year, which will help to update estimates for the asset and liability positions of small and medium-sized companies. The second is a household net worth survey that Statistics New Zealand will conduct during 2014/15. This will not give a complete picture, as it will not target trusts or high net-worth individuals. But it will be an important step forward in identifying a range of international assets held by New Zealand residents.

Some of the under-covered areas in the international statistics are also likely to correct themselves over time. For example, the upcoming introduction of a trans-Tasman retirement savings portability scheme will help over time to bring New Zealanders’ overseas pension assets currently held in Australia (ie, out of the reach of existing surveys) into Kiwisaver providers in New Zealand that are well covered by the current framework.

However, given the trade-off between improving data quality and increasing the reporting burden on survey respondents, other improvements will increasingly target administrative data (such as that from the IRD on income flows) and sharing more data with other international statistical agencies (so-called mirror-data).

Overall, given that offshore investments of individuals and trusts are likely to constitute the most significant gap in the measurement of our international assets, this will be a key area for future development work in the official statistics.


Page top