2 Methods and data
“There is no one single 'correct' definition of saving. The definition of saving used is that which is most appropriate to the issue under investigation”
Crossley and O'Dea (2010), p.37.
In this section we briefly outline three different approaches to the measurement of saving, and review the sources of data.[3]
2.1 The flow measure
Savings can be measured as a flow, defined as the difference between income and consumption.
SF = Y − C (1)
where:
SF = saving as measured by the flow measure
Y = disposable income
C = final consumption
Challenges arise in deciding exactly what should be included in the measures of income and consumption. These are addressed in Section 4.
2.2 The stock measure
A pure economic view of savings is the change in the stock of net wealth. The development of this approach is attributed to Haig (1921) and Simons (1938). Conceptually this can be applied at any level: to households, governments or nations. In practice the limitations of existing data systems means that estimates are typically restricted to households. The stock measure[4] is defined
Ss = ΔNW = SF + revaluation + other (2)
where:
Ss = saving as measured by the stock measure
ΔNW = change in net wealth
revaluations = the revaluation of real and financial assets
other = other changes in net wealth, such as transfers from other sectors, loss or destruction and discovery.
The flow and stock measures are conceptually equivalent after removing the effects of revaluations in real and financial assets, and other changes in net wealth arising from transfers from other sectors, discovery and destruction. In practice, data limitations mean it is very difficult to fully reconcile these two approaches. However, the net wealth approach remains a useful alternative to the flow measure.
2.3 An equity injection approach
Hodgetts, Briggs and Smith (2006) draw on an approach used by the Federal Reserve Board of Governors in the USA for estimating detailed flow of funds accounts. Saving can be derived from the net flows into and out of household assets and liabilities. They use aggregate data net acquisitions of financial assets (eg, currency, bank deposits, unit trusts, superannuation funds, direct purchases of equities) and net investment in tangible assets (eg residential fixed investment and land) to derive total net investment by households. From this they subtract net increases in financial liabilities (mortgages and other loans) as well as the consumption of fixed capital (depreciation) and net capital transfers.[5]
2.4 Data sources
In this section we review the available sources of data for measuring savingin New Zealand; Table 1 sets out those sources.
| Flow Approach: Income less Consumption |
Stock Approach: Change in net wealth |
|
|---|---|---|
| Micro economic data (eg household) |
Household Economic Survey |
SoFIE |
| Macroeconomic aggregates | System of National Accounts: Institutional Sector Accounts Household Income and Outlay Account |
Reserve Bank of New Zealand: Household Financial Assets and Liabilities and Housing Values |
Notes
- [3]For a comprehensive review of studies which estimate household saving rates based on a range of approaches see Browning and Lusardi (1996).
- [4]For a discussion of the measurement of net private wealth and the associated implied saving rate see Australian Treasury (2000).
- [5]We have not pursed this approach in the current paper; it remains for further research.
