Fraser Institute on Stimulus: Take Two

Iglika makes several cogent, high-level criticisms of the Fraser Institute’s “analysis” of how much government stimulus has contributed to Canada’s economic recovery. However, I think that it is guilty of a far more basic flaw.

To determine how much government purchases and investment contributed to economic growth, one would compare the increase in government purchases and investment with the increase in total economic output. A couple of weeks ago, I did those calculations using Statistics Canada’s Gross Domestic Product (GDP) figures.

These figures show that government purchases and investment, which accounted for only one-quarter of the economy in the second quarter of 2009, have accounted for one-third of the economic growth since then. How did the Fraser Institute reach a completely different conclusion using the same numbers?

The answer is that it did not examine stimulus as a share of economic growth. Instead, it compared the rate of increase in stimulus between quarters (which is also interesting, but in no way supports the Fraser Institute’s conclusion.)

A summary of the Fraser Institute’s tables follows (Private Sector = Private Consumption + Business Investment + Inventories + Net Exports). As was the case with my calculations, the components of GDP do not precisely add up to the total.





Gov. Purchases 




Gov. Investment




Private Sector




Total GDP   




In the second quarter of 2009, increased government purchases and investment partly offset reduced private-sector output, mitigating the recession.

In the third quarter, a further increase in government purchases and investment outweighed a further reduction in private-sector output, tipping the overall balance from recession to growth.

In the fourth quarter, the private sector returned to growth as yet another increase in government purchases and investment added to it.

However, the Fraser Institute’s “methodology” is to subtract each item’s contribution in one quarter from its contribution in the next quarter. Since stimulus contributed about the same amount to growth in each quarter, the Fraser Institute concludes that it contributed almost nothing.

Of course, it is true that government purchases and investment increased by similar amounts in each quarter. Meanwhile, the private sector swung from a large decrease to a small decrease, to an increase. But that is exactly how stimulus is supposed to work: steady growth in government purchases and investment stimulates a return to private-sector growth.

Anyway, I guess one has to grudgingly give the Fraser Institute some credit for translating month-old GDP numbers (released March 1) and meaningless arithmetic into extensive media coverage.

UPDATE (April 14): Mark Sandilands had a great letter in The Lethbridge Herald:

Fraser Institute take on stimulus plan off target

Readers must have been surprised to see the right-wing Fraser Institute attacking the federal government stimulus program (“Stimulus Plan Didn’t Help Economy,” Lethbridge Herald, March 31). After all, doesn’t the Fraser Institute usually cheer for the Harperites? Well, they do except when Harper and company are forced into doing something they don’t want to do by the majority of members of Parliament.

Memory takes us back to the fall of 2008 — the first time Harper prorogued Parliament to avoid a nasty situation: the confidence motion that might have toppled his government. The Tories went into overdrive, convincing a slight majority of Canadians that this was an illegitimate, undemocratic move by the opposition parties. Harper tossed away his credibility, particularly in Quebec, by accusing the Liberals and New Democrats of colluding with separatists. (All this is grippingly detailed in a recent book: “How we almost gave the boot to the Tories” by Brian Topp). The culmination of the drama was the budget of January 2009, brought in with enough stimulus to ensure that Michael Ignatieff and the Liberals would not vote it down.

So, did the stimulus plan help the economy? Fortunately other economists have analyzed the report done by the Fraser Institute and found it lacking. One (Erin Weir of the Progressive Economics Forum) analyzed Statistics Canada data and found that “government purchases and investment, which accounted for only one-quarter of the economy in the second quarter of 2009, have accounted for one-third of the economic growth since then.”

Weir then asked how the Fraser Institute got it so wrong. The answer is that the Fraser authors did not examine stimulus as a share of economic growth. Instead, according to Weir, they compared the rate of increase in stimulus between quarters. This would almost guarantee that no effect of the stimulus would appear. But then, that would suit the relentlessly anti-government-action Fraser Institute to a T, wouldn’t it?


  • Well said, Erin. I was trying to get at that in my response to Duncan’s comment on my post but you’ve done it much more eloquently. The government stimulus provides a large positive contribution to GDP that is stable from quarter to quarter, just like it’s supposed to, so that it can support the recovery of the private sector.

    The Fraser Institute report attempts to discredit the effect of the stimulus by asking the question “did the stimulus drive the spike in economic growth we observed in the second half of 2009” and arguing that it didn’t. Growth rates changed while stimulus spending was stable, therefore growth rates cannot be explained by stimulus spending. Completely missing out on any potential indirect effects of providing stability and putting a floor on job and income losses.

  • On the extensive media coverage front, check out the response from PM Harper in the Globe and Mail here.

    Instead, Stephen Harper said in launching a passionate defence of his stimulus spending, the criticism by the Fraser Institute was ideologically motivated by those opposed to government job creation.

    “I happen to be a believer that under normal circumstances, it is the private sector that creates jobs and it is going to have to be the private sector that will ultimately create long-term jobs,” Mr. Harper said in London, Ont., on Thursday.

    “But we are in the circumstances where [government spending] was required and when that is required, we are not going to act on the basis of ideology, we’re going to act on the basis of what the economy needs.”

    To summarize: FI is ideologically motivated. I happen to share their ideology, but I thought the recession wasn’t a time to act on ideological arguments.

    Hey, at least he’s honest.

  • Before writing all us Conservatives off as kooks.

    Comment from Brandon L
    Time: December 15, 2009, 3:41 pm

    “@David. There is no way, that had government not acted we would be existing recession but at the same time their would be 0% chance of malinvestment, misalocation of resources and illusory price gains. I wish Sir Keynes was still alive because his philosophy has been so perverted, that proponents seem to use it arbitrarily to fund perpetual wars & wasteful welfare.”

    “For an empirical link to more on base with your point. Look at how George bush inherited a rescession from a collapsing asset bubble in tech stocks followed standard economic theory often used by creditors. I challenge anyone to read Keynes collected writing on debtors. Bush/Greenspan lowered interest to 1% & ran deficits. Had Bush not engaged in standard economic theory then the rescession late 90’s would have been worse in hindsight their would have never been a housing bubble. Disappearing lending standards with declining saving rate in all ages is because of monetary policy. Price controls & capital controls have never worked long term.”

  • Here is a slightly different take on how to look at the GDP numbers.

    What the Fraser methodology glosses over is the fact that private sector driven upticks towards the end of the recession were from extremely depressed levels, and that the overall level of GDP going into the end of 2009 would have been significantly less if it had not been for government stimulus measures.

    This emerges clearly if we look at the change in the level of real GDP between Q4 of 2008 when the recession began, and Q4 of 2009.

    Over this period real GDP fell by $15.9 Billion or 1.2%. Meanwhile government current spending rose by $11.6 Billion or 4.4% and government capital spending rose by $7.75 Billion or 16.7%. The total increase in government spending of over $19 Billion clearly made a huge difference to the level of GDP in Q4 2009. Without it, the downturn would have been more than twice as large. (Though not all of the increase in spending can be counted as stimulus, most of it was discretionary.)

    Meanwhile, between Q4 of 2008 and Q4 of 2009, exports fell by $35 Billion or 7.5%, and non residential business investment fell by $31.4 Billion or 16.7%.

    My take is that we had a collapse of exports and business investment – offset to a hugely significant degree by government spending. (And government EI spending also supported personal spending.)

  • Alex Karaivanov

    I find both analyses flawed. Two cautionary comments:
    1. on the public sector jobs created. Self-proclaimed “progressive” economists always seem to attach an infinite value to government created jobs in a recession. As just a normal economist I would like to see some numbers on the cost side too – e.g., how much “stimulus” money was needed per such job created and what is the value added of such a job in comparison?

    2. The other “fetish” of proponents of (huge) government spending in recessions is the GDP and how much it falls in a given year. Of course, simply mechanically, one can bump up G in any given year by borrowing heavily from the future. Heck, one can even increase G (and hence GDP, by definition) a lot by selling off capital and spending the money. Instead, I would like to see a serious (dynamic, not just static) welfare analysis of the impact of the stimulus, not an accounting exercise of how much G or employment can be grown artificially. To give you an example – if GDP growth through gov’t spending and low unemployment were the sole criteria we should care about, the Soviet Union must have been Nirvana.

Leave a Reply

Your email address will not be published. Required fields are marked *