The Business Recovery Playbook USA and Canada’s Path Forward

Although a break at the middle of the series is arbitrary, tests for parameter stability using unknown break points are covered below. GMM with Newey and West (1987) robust standard errors (the lag truncation value is set to four) computes the estimates of the means, standard deviations, dynamic correlations and related standard errors.TripleWith equal volatility in these growth rates (standard deviations are little over 2 per cent), Canadian and US GDP have, on average, increased at similar rates, 3 to 3.5 percent. As is commonly known, the two economies have rather substantial contemporaneous correlation—0.74. This is high, relative to most other ties between developed nations. For the 1972–2000 period, Otto, Voss, and Willard (2003) for example find a mean correlation of 0.32 for the bilateral ties between 22 Organisation for Economic Co-operation and Development (OECD) economies. The dynamic correlations imply a somewhat greater association between lags of US GDP and current Canadian GDP than the reverse, therefore offering meager proof that the US economy leads the Canadian economy. This is in line with the theory that, at least in part due of their close trading ties, the US economy is a major external shock source for the Canadian economy.

Note, however, that given the related standard errors, the variations in correlations are. 

Not especially significant and, in this sense, it is difficult to build a compelling case in this regard. Fascinatingly, this theory that the US economy leads the Canadian economy does not fit the data from more official business cycle dating systems, such those offered by the Economic Cycle Research Institute (ECRI), which employs the same dating technique for several nations as used by the National Bureau of Economic Research (NBER) for the United States.5 We include a synopsis of the Canadian and US ECRI dates in Figure 2 as, taken from the standpoint of these business cycle dating approaches, several other fascinating points of comparison between the US and Canadian economies appear. One appealing aspect of the dating methods is that they let one compare across a broader sample period— 1948– 2004.Ten times the number of recessions (quarters between peak and trough) the United States has undergone compared to Canada reveals first noticeable differences. From this vantage point, a recession in the United States is insufficient cause for a recession in Canada, a conclusion not probably shared by many analysts on the state of the Canadian economy. Nor is this conclusion an artifact of a certain time period: idiosyncratic recessions in the United States emerge all across the sample, including the most recent downturn in the United States in 2001.Five The second startling aspect is the contraction time. There is a fairly coincidental contraction in the United States for all four contractionary phases Canada goes through. And in every case the Canadian economy set the benchmark.

This seems to go very counter to the idea that Canada is highl.Using parameter stability tests.

For GMM estimators suggested by Andrews (1993), we then investigate the instability of the moments of the output-growth series more formally. These tests have the benefit of letting one predict the timing of any structural break in the parameter estimation and of tolerating unknown break locations. Table 2 reports the results for these tests—more especially, the supremum of a series of standard Wald tests for equivalency of parameter vectors across a run of subsamples—as well as the corresponding break dates when pertinent. Also included are subsample estimations predicated on the projected break date. Analyzing the stability of the correlation coefficient for the two growth rates is the key goal. Still, this is not an easy chore. Other studies including Debs (2001) and McConnell and Perez-Quiros (2000) as well as the findings in Table 1 indicate that the means and standard deviations of the two series are not likely to be stable. Every test of parameter instability for the moments of these series will have to allow for this source of instability in addition to probable instability in the correlation coefficient. Setting up the estimation of the five-moment parameters (means, standard deviations, and correlation coefficient) concurrently and test for stability over the whole vector of estimated parameters is one easy approach of doing so. Although this is the initial strategy followed, it has two disadvantages. It first treats only one common date for structural modification over the whole parameter vector. Second, it just addresses the stability of the whole parameter vector; it does not offer direct information about the correlation coefficient itself. Another approach is to go in stages: first, find any structural change in the means and standard deviations; then, apply this knowledge to standardize the data so that.

Of all, a basic comparison of business cycle dates leaves out information on the links.

Between the economies at all points in the sample, not only the beginning of recession. For instance, fluctuation in US output is almost likely going to have some impact on Canadian output during times of US expansion. This is the reason the time-series characteristics of cyclical macroeconomic series take front stage here. Still, Figure 2 is instructive in that, at a broad level, the US and Canadian economies can function rather clearly.As Table 1 shows, the two economies have a somewhat close association during the previous forty-odd years. This is well known generally. Less is known about the changing nature of this link. Table 1 shows approximations from arbitrary subsamples, 1963Q1–1979Q4 and 1980Q1–2003Q1 (around the middle of the sample). Growth rates fall for these subsamples between the early and later periods; the dip in Canadian GDP (from about 4.5 per cent to less than 3%) is the most pronounced. Furthermore, there are differences in volatility: Canada's standard deviation of growth rates increases while the United States's falls. These findings will be examined in more detail in later detai and line up with past investigations. the means and standard deviations are constant over the whole sample. One can thus test the correlation coefficient's parameter stability by itself.Canadians relies on the US economy and that downturns in the United States lead, via trade and maybe other channels, to downturns in Canada. Otherwise said, Figure 2 shows no convincing data indicating the United States is a major cause of shocks to the Canadian economy.