Mar 24, 2010

The role of stimulus spending in the recovery


Yesterday, the Fraser Institute published a new report, which argues that the government stimulus did not drive Canadian economic growth in the last two quarters of 2009, suggesting that government spending on infrastructure was useless for the economy.

The report earned the scorn of Federal Finance Minister Jim Flaherty, who called it “poorly done and it’s wrong” in the Vancouver Sun.  His overall assessment?

“You know I just — I’m disappointed that the [Fraser Institute’s] work is as shabby as it is.”

Let’s take a closer look at the actual report and examine the claims made and the statistical evidence used to back them up.

To come to the conclusion that “government investment did not contribute to the turnaround during this period [the second half of 2009],” the Fraser Institute researchers engage in a simple growth accounting exercise, breaking down the change of GDP growth in the last two quarters into its four components. Every first year macroeconomics student would be familiar with the growth accounting equation:

GDP = G + I + C + (X-M)

In plain English (for the non-economist readers), this translates to:

Total economic output = Government spending (or consumption) + Investment (both government and business) + Private consumption + Net exports (the total value of all exports minus the total value of all imports)

Statistics Canada measures and reports on each of these four components on a monthly basis, and it also compiles the data into quarterly and annual reports. The Fraser Institute researchers looked at the numbers for the last three quarters of 2009 and found that government spending and investment played a relatively small role in Canada’s economic recovery so far.

On the surface, this seems to make sense – Canada is a small open economy and global conditions (such as commodity prices and demand for our exports) drive our economic fortunes to a large extent. Global forces started the recession and we cannot sustain a strong domestic recovery without an improvement in the global economic conditions.

In fact, I agree with the Fraser Institute’s assessment that the government cannot claim exclusive credit for the speed of the recovery, just like it cannot be blamed for the recession in the first place. Along those same lines, we should acknowledge that the BC boom in the late 2000s was driven by high commodity prices and growing demand for our exports, rather than by the changes in domestic tax rates or the government’s economic management (although you won’t hear the Fraser Institute staff make this argument).

However, while the government did not single-handedly drive the recovery, they can and should claim credit for cushioning the blow of the recession. Both the federal and provincial governments stabilized employment and personal incomes (and thus consumer spending) by maintaining operating expenditure levels despite the reduction in their revenues. The federal stimulus package was arguably too small to make a very big dent and there were delays in rolling out the actual spending, but without it things would have been a lot worse. Note, for example, that almost all of the recent gains in employment have come from the public sector, while the private sector continued to shed jobs in February, according to Statistics Canada’s latest release from the Labour Force Survey. How much higher would unemployment rates have climbed if the government listened to the Fraser Institute’s advice to balance their budgets by cutting spending?

Government spending filled part of the gap opened by the decline in business investment and private consumption during the recession. With public spending at less than a third of Canadian GDP, the government cannot be expected to fully offset the drop in business investment or private consumption. Their job is to smooth out economic fluctuations by spending counter-cyclically and this is precisely what they did this time. The Fraser Institute researchers do not find evidence of that because they do not look. They focus their attention exclusively on the last two quarters of 2009, when recovery is starting to take hold.

There are a number problems with their analysis.

1) They misuse the growth accounting equation to derive conclusions that are beyond the scope of the equation. All the growth accounting equation tells us is how individual GDP components changed each period (i.e. did they rise or fall), it does not tell us why.

2) They assume that contemporaneous correlation equals causation (and this one is a classic concern in social sciences). The Fraser Institute reports looks at what variables change between two quarters and argue that these must be causing the observed change in GDP growth. This simplistic analysis neglects indirect effects that the stimulus spending may have had on consumer and business confidence, for example. The report authors also fail to account for a number of other government interventions which could conceivably explain some of the GDP growth, such as the swift monetary response from the Bank of Canada, which lowered interest rates and eased credit conditions undoubtedly contributed to improving the climate for private investment.

3) Two quarterly observations do not provide sufficient data to derive meaningful conclusions about the drivers of economic growth in BC, particularly so when the numbers fluctuate so widely from quarter to quarter (compare figure 2 and figure 3 in their report).

Business investment contributed 1.0 percentage points to GDP growth between the 2nd and 3rd quarter, but -0.6 percentage points between the 3rd and the 4th quarter, offsetting the previous quarter’s gains by more than half. Similarly, net exports were a drag to the economy in the 3rd quarter but a boost in the 4th. Which way these numbers will go next quarter only time can tell.

Looking at annual data provides very different conclusions. Statistics Canada reports:

Real GDP fell 2.6% in 2009, as exports dropped 14% and business investment in plant and equipment fell 17%.

and later on:

Final domestic demand recorded a 1.7% decline, largely due to a 17% drop in business investment in plant and equipment. Consumer spending edged up, as a result of a 1.1% increase in expenditures on services. Consumer spending on durable and semi-durable goods declined 2.8% and 2.9% respectively. Government current expenditure on goods and services grew 3.0%, and government capital expenditure increased 13%. [empasis added]

It’s clear from the annual data that without the government stimulus spending, the economic decline in 2009 would have been far greater. It’s quite possible that government stimulus spending provided the necessary conditions for increased private sector spending, thus indirectly driving economic growth. For example, both the federal and provincial governments made it very clear that they are prepared to step in and avert the worst consequences of the recession even at the cost of budget deficits. This certainly helped restore business and investor confidence and opened the door for private sector investments that may not have been undertaken otherwise.

The Fraser Institute report finds that net exports were the main economic driver in Canada during the 4th quarter of 2009. Unfortunately, this wasn’t the result of a strong rebound in our export sector – exports continue to suffer from the high Canadian dollar – but of “a slowdown in the growth of imports of goods and services.” Lower imports means weaker domestic demand. This is hardly the foundation for a strong private sector recovery brewing.

Indeed, it’s too early to conclude that the private sector would be able to take off on its own should the stimulus be withdrawn. We need to be very careful about scaling back public spending next year to avoid slipping back into a recession.

Topics: , ,