Donald Trump’s adviser Steve Bannon once claimed US conservatives were going to “go crazy” about the new administration’s infrastructure plans. Trump had spoken endlessly about investing $1 trillion (£773bn) in American highways, roads, bridges, tunnels, airports, and more, and how this was going to create jobs and improve growth.
This won praise even from his Democrat opponents, but coming so soon after President Obama’s 2009 “stimulus” programme, was the stuff of fiscal conservative nightmares.
Fast forward to today and it’s the Democrats upset with Trump again. The one issue which offered the possibility of cross-party co-operation has again broken along partisan lines.
That’s because in his budget a fortnight ago, Trump relegated infrastructure to his second tier of priorities, pledging a much smaller $200bn (£155bn). Instead of federal spending, Trump’s team wants to focus on reducing regulations, streamlining permitting processes and encouraging user fees, such as tolling, to boost private investment to get to the $1 trillion total. To Democrats, who want more spending and are concerned by more private sector involvement in infrastructure, this is seen as a missed opportunity to boost growth significantly.
But is it? The Democrat position has become conventional wisdom among commentators. More government infrastructure spending is seen as a “win-win” that can boost GDP in the short term, creating high-paid jobs, and enhance productivity in the longer term, through better connectivity and mobility. Just last week the OECD called for more public investment in the UK.
Yet there are good reasons to think both claims are severely overblown, particularly for the US.
In fact, now is probably the worst possible time for a fiscal stimulus there. The unemployment rate has fallen to 4.3pc, below the level that many organisations believe is sustainable. There are some reports of labour shortages. In such an environment, more government borrowing for major projects will simply shift workers from private sector activity into government-driven activity. This is particularly true given the recorded unemployment rate in construction is now at its lowest level since the height of the boom in 2007.
More government borrowing in a period when interest rates are rising and with full employment will also increase inflation, leading to higher interest rates more quickly, in turn discouraging capital investments in the private sector. Increasing federal spending temporarily will do little to boost short-run GDP but will raise US public debt further.
No serious economist, of course, would argue that having good infrastructure in the long term would not enhance an economy’s growth potential. With productivity growth projections in the US so low compared to historic rates, one can see the appeal of thinking careful, targeted, long-term infrastructure investment could really help boost growth. But the real question is whether a centralised national infrastructure programme is the best means of achieving good infrastructure and higher productivity.
Japan threw $6.3 trillion at infrastructure between 1991 and 2008 and, while they have an excellent rail system and other marvels, it did little to boost long-run growth. A recent academic study of major Chinese infrastructure investment projects undertaken by the University of Oxford found that over half had benefit-to-cost ratios below 1 — net losses in economic value.
When resources are allocated through the political process, many considerations other than growth come into play. Politicians do not choose the unsexy but crucial projects such as road maintenance, but prefer megaprojects. In the UK, the 2010 spending review allocated funds to HS2 even though the benefit-cost ratio is around 1, deferring or cancelling road schemes with ratios of 6.8 and 3.2 respectively.
In the US, research has found that allocating federal funding for highways on the basis of benefit-cost ratios could deliver the same outcomes for the system at 30pc less cost.
In other words, though theoretically governments can enhance growth by investing in good projects with higher returns than the private sector, in practice they do not. The book Megaprojects and Risk by Bent Flyvberg and others at Oxford University has shown that governments’ conflicting role of both promoting and being responsible for projects often leads to a significant bias toward overoptimism. In a study of 258 projects across 20 countries, they found nine out of 10 ended with cost overruns.
Given all this, Trump should ignore Democrat whining. He is surely right that re-examining the incentives, frameworks and institutions surrounding infrastructure provision are more important than merely chucking more money at it. The US certainly has significant infrastructure needs given areas of high congestion, but one could easily see a huge amount of taxpayer money being wasted on bad investment.
A thought-through supply-side agenda which cut red tape and removed barriers to privatisation could make infrastructure development both more responsive to demands and cheaper for US taxpayers. The devil will be in the details, but Trump is right to reject the principle of a mammoth infrastructure spending spree.