Infrastructure Spending Stimulus: Good in Economic Theory – A Fail in Political Economy Practice?

For every complex problem, there is a solution that is simple, neat, and wrong.
– HL Mencken

‘Canada’s national newspaper’ the Globe and Mail on January 16th 2016 (Folio infrastructure A8) quoted a ‘simple’ Finance Canada table showing that  for every dollar spent on infrastructure there is an increases in economic growth by  1.5 dollars. The article quotes the Department as concluding that  “infrastructure spending is the most effective form of stimulus compared to the other options,  including more generous Employment Insurance benefits or tax cuts.”   Private sector economists apparently agree with the idea that infrastructure spending stimulus would be significantly positive – boosting economic growth by an estimated half a percentage point.

The problem is that we have been here before, and the  evidence suggests that there is a fly in the ointment. In late 2010 Tim Kiladze of Canada’s national newspaper noted that Economic Action Plan put in place by the government of Canada in 2009 after the 2008 recession for mostly so-called ‘shovel ready’ infrastructure  funding was spent too late to be effective. In fact, it was suggested by some analysts he quotes in 2010 – that some infrastructure funding in 2009 was actually delayed while provinces and regions waited to see if their project qualified for federal assistance. See So the Economic Action Plan Infrastructure spending of 2009-10 appears to have potentially de-stimulated the economy in some areas during the Canadian Economy’s  time of greatest stimulus need! (The systems thinkers call this a classic ‘fix that fails’.). The real problem from my perspective  is that some of us predicted this. (See

How can we use evaluation expertise to help this time? The first step will be to recognize the problem here. I submit that the implementation design did not fit the need and the theory of change in 2008-09. In simple terms – there were too many ‘partners’ and intermediaries who needed to align to spend the cash. Multiple levels of government were involved, diverse communities and various civil society stakeholders and the media were weighing in on almost each and every investment. By contrast – the US and then Canadian equivalent of cash for clunkers used straight forward authorities (i.e. The programs featured direct federal government funding to intermediaries.) and motivated intermediaries who had a track record of speedy transaction promotion  (i.e. car salesmen) to push federal stimulus money out the door at record levels.  See for a more detailed discussion and some admitted gloating about being right in my 2009 prediction that the Economic Action Plan infrastructure stimulus would not work.

Hopefully this time we can get our Canadian Federal Government to consider the linkage of the theory of change (i.e. the need to stimulate the economy through some transfers) with the implementation design. The evaluation and research community then needs to step up and to synthesize the conditions and contexts that will enable stimulus programs, such as those being contemplated right now, to succeed in the political economy of Canada. We need to help fill in the ‘gap’ that frequently exists between the policy theory and the implementation reality. In some of the latest thinking on the use of theories of change from Mayne ( – see especially pages 120-122)  the point is made that program theory should be used for much more than the design of evaluation or measurement regimes. When it includes concepts like who one needs to reach and what early and intermediate changes need to occur – it  can be used to directly inform policy and programming. Isn’t it time  to get in front of the big decisions – so evaluators and evaluative thinking can help inform policies , programs and strategic decisions? For me that would be way more fun than toiling away on our studies in obscurity and/or continuously saying “I told you so” after the fact.