In this budget the Treasurer’s speech referred to the government’s “$110 billion investment pipeline” and announced new infrastructure initiatives worth $15 billion. Further documentation claims the new spending will support “over 30,000 direct and indirect jobs across the lives of the projects”.
Last year’s budget and earlier measures during the COVID recession also stressed infrastructure spending. Over the course of 2020 the government announced a number of stimulus measures to offset the impact of COVID-19 and the lockdowns that followed. This suggests a good opportunity to look back at what has been achieved so far.
High unemployment presents us with a very good opportunity to undertake projects that will provide benefits into the future. Calculations using figures in last year’s budget suggest the first $100 billion was only going to produce one job (direct and indirect) for every million dollars spent. The language is vague, so we do not know if these are total jobs or full-time year equivalent jobs, or if multiplier effects are included and so on. What we can say though is the impact is small in terms of bang-for-buck measures. It is not clear what is included in the further 40,000 jobs and whether they are net figures, or some brought forward and so on although the suggestion is that the bang-for-buck is somewhat more effective. On top of these measures there were reports the Federal Government was also urging additional stimulus on the part of the states and territories.
Given all this we might expect to see national spending on government infrastructure showing a large increase. However, that is not so apparent when we examine the actual data for general government “gross fixed capital formation” or “investment” in plain English. Figure 1 shows government investment over the last 13 quarters from December 2017 to December 2020. It also includes total public sector investment which includes investment on the part of government business enterprises. All data is expressed in chain volume measures.
The data illustrated in Figure 1 suggest that since December 2017 government investment had been increasing but, if anything, it has slowed down since the December 2019 quarter although there may be the hint of an uptick in December 2020. The actual figures show an increase in general investment of $684 million which is small in the context of the billions being talked about by the government in the budget context and elsewhere. Compared with the previous December quarter the increase hardly registers at just $96 million. The figures for general government investment suggest that any new spending has been offset by cuts or reductions in other capital spending. Governments may be tempted to reduce spending in other areas to finance new initiatives but that of course defeats the impact of the new initiatives. Yet that seems to be what has happened. The Reserve Bank minutes of its April 2021 meeting noted that “Indications at the time of the meeting were that the rollout of public investment programs over the first half of 2021 would be slower than foreshadowed in state budgets”.
Incidentally, it is worth noting that most Commonwealth Government capital initiatives do not show up as such in the budget papers. Usually, the Commonwealth provides grants to the states for spending on infrastructure in which case the capital spending shows up in the state and territory books while the relevant grants are treated as current outlays in the Australian Government budget documentation.
What do we conclude from all this?
We have made the point that the traditional stimulus package focused on construction is not a very good jobs generator and fails to address female unemployment. But of course, as Keynes made clear years ago, poorly chosen public works are better than nothing so long as they increase employment. However, since the data suggest there is no net increase in investment all of this is moot.
The evidence to date is that government initiatives on infrastructure spending are not significant enough to show up in the national accounts as net increases in government capital spending. In order to be effective at the national level new public works should represent a net increase in total spending on public works and evidently that is not the case.