The measurement and implications for economic growth of government capital stock in a small open economy: New Zealand 1950-90

Abstract

This thesis is concerned with the production of capital stock estimates for the government non-market production sector for the period 1950 to 1990 via the Perpetual Inventory method. Results of the literature review on the quantitative impact of government expenditures discussed in chapter two suggest a case for a relationship between government investment and economic growth, and while the chief motive for estimating government investment variables is to add to New Zealand's historical statistics the assembly of government investment time series can allow similar tests of the relationship between a clearly defined government test variable and economic growth to be undertaken for the post-1949 New Zealand economy. Chapter four is concerned with the production of capital stock estimates for the government non-market health, education, and defense and administration sectors. Chapter five deals with the road and harbour works sector which is considered to require special treatment given ambiguities on the survival life of the assets under the former category. Emphasis is placed upon the production and assembly of gross capital formation estimates, an important requisite for the perpetual inventory model, as far back in time as the formation of the oldest capital item expected under certain service life assumptions to be still in existence in the capital stock at 1950. While error in the capital formation series is likely to be greatest with estimates furtherest back in time, this become progressively less important through time to the final capital stock estimates. Where possible final stock estimates at benchmark dates are compared with estimates produced using different and often incomplete data sets and/or those derived under different estimation procedures. Possible explanations are given where significant deviation between estimates is obtained. The pattern of non-market capital stocks and capital formation for New Zealand for the period 1950-90 is considered in chapter six. Where data is available the New Zealand experience is compared with that of other OECD nations, such a comparison being extended to the area of general government investment allowing a comparison between a greater number of OECD members. The pattern in other government variables within the OECD is also considered, as under budgetary constraints it can be considered that such variables will be one influence on the non-market investment variable. Some comments are made on the implications of the New Zealand experience in light of this comparison and in light of the results of the literature survey of chapter two. The literature survey of chapter two is concerned with quantitative tests of the impact of various government variables and economic growth, from which three dominant themes arise. The first is that it is most likely that government investment variables have a positive influence on growth, government consumption variables have a negative impact, and government transfers have an ambiguous influence, and given that the recent New Zealand experience as well as the recent experience for much of the OECD is for a reduction of the former and a rise of the latter two this is considered to be of some concern. The second strong conclusion is the need to consider real government test variables given a differing rate of inflation in the public and private sector. As relevant public deflators are only available for New Zealand for a part of the period of investigation further research work may be necessary before the non-market investment estimates generated by this thesis are to be utilised in regression tests similar to those surveyed in chapter one. The final conclusion centres around the ambiguity of the conclusions of the literature survey sample, which may or may not be a feature of the specification or lack of specification which underlie many of the tests surveyed.