By Steven Globerman and Paul Storer
Globerman and Storer update their statistical analysis of Canada-U.S. trade through 2012, and are able to identify “year-specific” trade impacts. The good news? The negative effects on Canada-U.S. trade stemming from 9/11 peaked in 2004. But Buy America provisions and a decline in the trade of intermediate goods have kept Canada-U.S. trade from returning to pre-recession peak levels.
In our statistical studies (Globerman and Storer 2007, 2009), we found evidence of significant declines in Canadian exports to the United States in the aftermath of border security measures implemented in response to the 9/11 terrorist attacks. While our 2009 study extended the end-point of our 2007 analysis from 2005 to 2007, we are now able to add data through 2012. Our statistical method identifies “year-specific” impacts of border security-related impediments to trade for individual years over the period 2002 through 2012. The additional data afford the potential to identify the impact of recent policy initiatives implemented to mitigate the adverse impact of border security on trade, including the Beyond the Border initiative.
When we include the additional years in our statistical analysis, the results remain largely unchanged for the immediate post-9/11 period. In particular, we identify statistically significant negative effects on Canadian exports to the United States in 2001 through 2003, after taking into account traditional trade determinants. After we add additional years, we still find that the negative impacts increase through the fourth quarter of 2001, a time when border security measures had been ramped up, and policies to mitigate the impact of security on trade were not yet widely implemented. The size of the negative effect gradually declines through 2003, as participation in programs such as FAST grew, and other operational and policy adjustments were made at the border.
After 2003, the only other years with statistically significant negative effects on Canadian exports were the recession years of 2008 and 2009. In fact, these recessionary effects are larger in size than our biggest post-9/11 impact prior to 2008 (we find 31 percent and 23 percent Canadian export shortfalls for 2008 and 2009). While we expect a decline in U.S. purchases of Canadian goods due to the recession’s impact on U.S. GDP, our statistical analysis includes U.S. GDP as an explanatory variable. Hence, our estimates of recessionary effects are beyond what would be expected due strictly to the drop in U.S. GDP. One possibility is that Buy America features of the U.S. government’s stimulus package may have caused a substitution away from Canadian products in favor of U.S.-made products.
A useful illustration of the relationship between Canadian exports and U.S. GDP is provided in the figure below which shows Canadian exports to the United States (in U.S. dollars) as a fraction of U.S. GDP (also in U.S. dollars). We see an initial moderate decline in this ratio in the late 1980s and early 1990s, as Canadian exports to the United States grew more slowly than U.S. GDP. Afterward, exports generally grew more quickly than U.S. GDP, with a few exceptions such as the immediate post-9/11 period. During the recession, the ratio of exports to GDP declined rapidly to roughly its 1991 level. By mid-2012, the ratio was still well below its peak pre-recession level.
The impact of the recession on Canadian exports to the United States could exceed the historical direct impact of declines in U.S. GDP, because much of Canada-U.S. trade is two-way trade that often occurs within an industry (in many cases this trade is between Canadian and U.S. operations of the same firm). Much of this two-way trade involves trade in intermediate goods, such as parts and accessories in the automotive industry. Another of our current research projects indicates that in 2009 there was a decline in the fraction of trade that comes from intermediate goods. Policymakers should be concerned about this drop in intermediate goods trade due to its importance for economic integration. The large and fairly persistent impact of the recession on Canadian exports to the United States is of general interest and probably merits further analysis. It could be that the automotive sector accounts for much of the observed decline in Canadian exports as a share of U.S. GDP.
In summary, our recent statistical results suggest that the negative impact of border security developments on Canadian exports likely peaked by the end of 2004. The behavior of Canadian exports to the United States as a share of U.S. GDP in the period 2008-2012 likely reflects the primary influence of other factors including Buy America U.S. government programs and cyclical declines in intermediate goods exports from Canada.