An examination of corporate tax cuts and growth in today's Globe and Mail notes that even though corporate income tax (CIT) rates have been falling since 2000, the share of nominal expenditures on investment in machinery and equipment of nominal GDP fell instead of rising.
Unfortunately, the analysis has a couple of problems that should be noted.
It’s important to make the proper distinction between prices and quantities. Nominal expenditures are the product of prices and quantities, and in this context, what matters is the quantity: the number of new machines installed matters more than what was paid for them. In addition to the nominal series used in today’s Globe, Statistics Canada also publishes data for ‘real’ investment spending, that is, expenditures corrected for fluctuations in prices. And it is these data that are used to calculate Statistics Canada’s estimates for productive capacity (pdf).
If you strip out price effects and look at the quantities, the trend is reversed; see the graph below. The explanation is simple: the relative prices of machinery and equipment were falling over this period. Since much of these goods are imported, firms could take advantage of an appreciating Canadian dollar to buy more investment goods at lower prices. What the graph in the article shows is that prices were falling more rapidly than quantities were rising – it does not mean that quantities were falling.
Policy evaluation problems
Even though real investment did grow more rapidly than real GDP since 2000, we shouldn’t go on to conclude that the CIT cuts were the cause. Such a conclusion would be based on the logical fallacy at attributing changes in investment to changes in the CIT rates simply because the policy change happened first. The correct way to look at the problem is to consider the counterfactual: what would have happened without the reduction in CIT rates? The difference between the counterfactual and what we actually observed is the effect of the policy.
Constructing a counterfactual requires addressing any number of modelling and statistical challenges; I’ve compiled a reading list of empirical studies here. Although modelling strategies and data sets vary from study to study, the consensus from the peer-reviewed academic literature is clear: lower CIT rates are associated with investment levels that are higher than what they would have otherwise been.
Follow Economy Lab on twitterReport Typo/Error