The Trump-Hammond-Keynesian consensus is wrong: Higher infrastructure spending does not equal higher growth
15 November 2016
SUGGESTED
Government and Institutions
14 November 2016
Economic Theory
16 November 2016
Government and Institutions
To a man with a hammer, everything looks like a nail. Or when it comes to the UK economy: to an economist who believed in fiscal stimulus in 2011, every period of time looks like an opportunity for more government spending.
In the aftermath of Donald Trump’s victory last week, even his most vociferous left-leaning economic critics were heralding the fact that he has advocated huge increases in infrastructure spending – something which Keynesian economists here have been calling for over much of the past six years.
In fact, this “increase infrastructure spending” clarion cry is fast becoming political consensus. Labour has argued for it since 2010. And Chancellor Philip Hammond has given some not-so-subtle indications that the Autumn Statement will contain provisions for more in the way of this type of spending.
It’s all a bit baffling. The Keynesian theory is that at times of high unemployment – when the economy is not running at its “potential” – temporary government spending on infrastructure (roads, bridges, rail) can increase demand and keep growth on a steady path. This spending can then be withdrawn as the economy recovers. In other words, governments can fine-tune the economy to smoothen the business cycle.
There are lots of challengeable assumptions to this theory, not least how any increase in spending might be offset by monetary policy and how consumers and investors might respond to the expectation of future higher taxes to fund the investment. There is also the little issue of most infrastructure projects having long time lags.
But even if one assumes these issues away, there is little evidence the UK is in some sort of heavy slump. Unemployment is very low at 4.9 per cent and the employment rate at historic highs. Data last week showed that UK firms hired permanent staff at the fastest pace in eight months in October. Retail sales are strong. The economy is growing at 2 per cent annualised (exactly the growth rate forecast in the 2016 Budget).
Lots of economists still reeling from the Brexit vote just assume the economy will be weaker next year, but these are the same economists who forecast weakness this year too – perhaps highlighting one of the key practical arguments against the use of this type of fiscal policy: that because of forecast errors it actually risks exacerbating booms and busts.
Advocates of more infrastructure investment put these critiques aside by saying that it is really a supply-side policy. With borrowing costs so low, they believe government investments can improve the long-term productivity performance of the economy in a cost-effective way. Investing in transport infrastructure can enhance economic mobility, for example.
They are right that good infrastructure can improve economic performance. But the international evidence that government infrastructure spending facilitates higher economic growth is surprisingly mixed.
The main reason can be summarised in one word: “politics”. For rather than investing in things with high economic returns, political choices often mean investment in favoured regions, or on projects with lower economic returns but higher political rewards (think HS2 rather than strategic road schemes with much higher benefit-cost ratios). Devoid of market pressures, state infrastructure projects are often beset by cost overruns.
Japan spent $6.3 trillion of public money on big projects, and ended up with bridges to nowhere. Spain, likewise, was left with empty airports. A recent assessment of a host of major projects in China showed a great deal of infrastructure investment there was plagued by overestimated benefits, meaning that 55 per cent of the projects had a benefit-to-cost ratio below one, i.e. they led to a net loss in economic value.
That’s not to say all government infrastructure investments are worthless, merely that political decisions are unlikely to result in a significant upward impact on economic growth, as promised.
If we really want to improve the UK’s economic growth rate, we have to weigh up the value of spending against other things the money could be used for (including growth-enhancing tax cuts). We also have to assess each marginal decision according to its economic merits. After all, just because rail may be essential for economic mobility, it does not mean every new rail project will bring economic gains.
In developing his Autumn Statement, the chancellor should consider these flaws and think through the means necessary and current barriers to private sector infrastructure investment as a priority. As the failure of post-Brexit forecasts shows, a consensus need not be correct – even if it unites Trump, Hammond and the Keynesians.
This article was first published in City AM.
In the aftermath of Donald Trump’s victory last week, even his most vociferous left-leaning economic critics were heralding the fact that he has advocated huge increases in infrastructure spending – something which Keynesian economists here have been calling for over much of the past six years.
In fact, this “increase infrastructure spending” clarion cry is fast becoming political consensus. Labour has argued for it since 2010. And Chancellor Philip Hammond has given some not-so-subtle indications that the Autumn Statement will contain provisions for more in the way of this type of spending.
It’s all a bit baffling. The Keynesian theory is that at times of high unemployment – when the economy is not running at its “potential” – temporary government spending on infrastructure (roads, bridges, rail) can increase demand and keep growth on a steady path. This spending can then be withdrawn as the economy recovers. In other words, governments can fine-tune the economy to smoothen the business cycle.
There are lots of challengeable assumptions to this theory, not least how any increase in spending might be offset by monetary policy and how consumers and investors might respond to the expectation of future higher taxes to fund the investment. There is also the little issue of most infrastructure projects having long time lags.
But even if one assumes these issues away, there is little evidence the UK is in some sort of heavy slump. Unemployment is very low at 4.9 per cent and the employment rate at historic highs. Data last week showed that UK firms hired permanent staff at the fastest pace in eight months in October. Retail sales are strong. The economy is growing at 2 per cent annualised (exactly the growth rate forecast in the 2016 Budget).
Lots of economists still reeling from the Brexit vote just assume the economy will be weaker next year, but these are the same economists who forecast weakness this year too – perhaps highlighting one of the key practical arguments against the use of this type of fiscal policy: that because of forecast errors it actually risks exacerbating booms and busts.
Advocates of more infrastructure investment put these critiques aside by saying that it is really a supply-side policy. With borrowing costs so low, they believe government investments can improve the long-term productivity performance of the economy in a cost-effective way. Investing in transport infrastructure can enhance economic mobility, for example.
They are right that good infrastructure can improve economic performance. But the international evidence that government infrastructure spending facilitates higher economic growth is surprisingly mixed.
The main reason can be summarised in one word: “politics”. For rather than investing in things with high economic returns, political choices often mean investment in favoured regions, or on projects with lower economic returns but higher political rewards (think HS2 rather than strategic road schemes with much higher benefit-cost ratios). Devoid of market pressures, state infrastructure projects are often beset by cost overruns.
Japan spent $6.3 trillion of public money on big projects, and ended up with bridges to nowhere. Spain, likewise, was left with empty airports. A recent assessment of a host of major projects in China showed a great deal of infrastructure investment there was plagued by overestimated benefits, meaning that 55 per cent of the projects had a benefit-to-cost ratio below one, i.e. they led to a net loss in economic value.
That’s not to say all government infrastructure investments are worthless, merely that political decisions are unlikely to result in a significant upward impact on economic growth, as promised.
If we really want to improve the UK’s economic growth rate, we have to weigh up the value of spending against other things the money could be used for (including growth-enhancing tax cuts). We also have to assess each marginal decision according to its economic merits. After all, just because rail may be essential for economic mobility, it does not mean every new rail project will bring economic gains.
In developing his Autumn Statement, the chancellor should consider these flaws and think through the means necessary and current barriers to private sector infrastructure investment as a priority. As the failure of post-Brexit forecasts shows, a consensus need not be correct – even if it unites Trump, Hammond and the Keynesians.
This article was first published in City AM.
1 thought on “The Trump-Hammond-Keynesian consensus is wrong: Higher infrastructure spending does not equal higher growth”
Comments are closed.
SHARE
One often forgotten element of infrastructure spending initiatives are their whole life service and maintenance costs. In the civils and construction industries a rule of thumb measure is that these can be six times the original capital cost. This may sound high, but today’s infrastructure has to be looked at from a systems engineering perspective; that means future service costs for service management, control equipment, data systems, fleet and personnel administration, replacement parts supply chains as well as the rather less expensive ten tons of asphalt to fill in potholes.
Where politicians choose “stimuli” based only on capital cost inputs, they ignore this forward overhead, which ends up on taxpayers’ shoulders through the years in a generally higher tax rate burden.
PFI attempted, rather clumsily, to contract-in these whole life costs, but it often turned out that the public servants signing the contracts were no match for industry specialists who knew how to manage their maintenance risks at taxpayer expense. There is also a tendency, especially at local authority level where capital spending and revenue spending are distinctly separated, for officers to bundle “some maintenance” into small-scale infrastructure contracts for a short period (less than six years). The time-horizon to the moment when servicing has to be re-negotiated is left long enough such that it “won’t be my problem” for the officer, while it is short enough that a top-up maintenance cover cost can be kept low enough to put forward as sensible. Such public choices incentives allow these projects to get the go ahead.
There is a case for a statutory requirement that a formal process in evaluating whole life costs is required, and those costs to then have to be agreed democratically for projects. There has been work by the Cabinet Office in this regard, but of course the political incentives are always that building monuments is more interesting and politically valuable than arcane calculations about maintaining electrical systems.
Sadly, taxpayers therefore do not get transparency over the imposition of these future stealth tax burdens. They simply emerge in the continual cry from the public sector that it needs more of our money to support their over-spent plans.