Neoclassical theories of economic growth emphasise investment in physical capital. For a country's economy to grow, its stock of capital goods - inventories, plus assets such as factories, machines, buildings and infrastructure - should be expanding.

While these theories are incomplete (and misleading, in that they underemphasise intellectual, social and environmental capital), statistics on expenditure on physical capital can tell us much about the state of an economy.

Economists call spending on fixed capital assets "capital formation" or "gross fixed capital investment" (GFCI). Thus, to economists, "investment" (which generally means spending by producing firms) is not the same thing as savings (which is done by households). Rather, savings, which means income not used for present consumption, are made available (generally but not always through the intermediation of banks) to firms for investment purposes.

It is important to note that 'investment' by the public in, say, the sharemarket, is, to economists, "savings" and not "investment". Likewise, foreign investment in New Zealand really means that foreign households are making their savings available to New Zealand located firms to use as investment capital.

In classical and neoclassical economics, savings by households is a prerequisite for investment. Hence, such economists emphasise the importance of savings to future economic growth. In Keynesian economics, however, investment is autonomous. It is determined by factors such as firms' confidence in future sales, and is made possible by new money, created as bank credit. In Keynesian economics, increases in savings are more a result than a cause of increased investment; and unemployment is the consequence of low rates of investment rather than of low rates of saving.

The above graph, based on quarterly current price data from Statistics New Zealand's "Expenditure on Gross Domestic Product" series, shows investment levels relative to GDP from 1982 to March 1999. "Total investment" includes increases (sometimes decreases, as in 1991) in firms' inventories (called, in the language of national accounting, "increase in stocks").

In recent years, the GFCI has been broken down into investment in housing and the physical capital purchases of firms, including State Owned Enterprises (SOEs).

The most significant conclusion from the graph is that, since the economic liberalisation that took effect from 1985, expenditure on investment goods has fallen relative to GDP. This is despite the fact that GDP per capita has been growing at below the post-war historical norm of two percent per annum (see "A Sound Economic Recovery?"). While the high proportion of investment in the early 1980s reflects the much criticised Think Big energy projects, the ratio of investment to GDP in that period is close to the post-war norm. Indeed, it was levels of gross fixed capital investment around 25% that sustained full-employment and 2% per capita growth rates from 1954 to 1976.

Another important feature of the recent investment statistics is that investment in housing has grown at a faster rate than has investment by firms. This is probably a direct result of the high interest rate monetary policy of the mid-1990s. Firms are more sensitive to high interest rates than are home purchasers. It is at times of high interest rates (and high exchange rates) that banks seek to expand their mortgage lending, and to reduce their exposure to their exposure to the tradeable sectors (especially farming and manufacturing) of the economy.

While investment expenditure in New Zealand in the 1990s appears to have been inadequate to restore New Zealand to full employment and to a growth trend of 2% per capita per annum, it is unlikely that an increase in investment spending and a decrease in consumption spending will, on its own, solve the problem. Of more importance than the quantity of investment is its quality. Furthermore, when most investment goods are imports, more investment may mean a significant worsening of New Zealand's chronic balance of payments problem. But that's another story.

© 1999   Keith Rankin

Rankin File | StatsWatch