AI Analysis: Government of Canada bonds outstanding, shown at par value, by currency of payments

Executive Summary

Government of Canada bonds outstanding grew nearly 38-fold over 51 years, from approximately $93.5 billion CAD in January 1975 to a peak of $3.56 trillion CAD in May 2019, with Canadian dollar-denominated bonds consistently dominating the portfolio. The dataset (Statistics Canada, Table 10100130) covers 614 monthly observations across 11 currency categories and 33 issuer types, revealing strong structural concentration in domestic currency issuance and a heavily right-skewed distribution where the median value of $187 million CAD sits far below the mean of $25,320 million CAD. Notable anomalies include 1,319 negative-value rows isolated to the 'Other currencies' category and a cluster of 38 statistical outliers concentrated around mid-2016, suggesting a structural shift or reclassification event during that period.

Key Findings

• Total Government of Canada bonds outstanding grew from ~$93.5 billion CAD in January 1975 to a peak of ~$3.56 trillion CAD in May 2019, representing a nearly 38-fold increase over 44 years.
• Canadian dollar bonds and total currencies are nearly perfectly correlated (r=0.995), confirming that CAD-denominated issuance overwhelmingly drives the overall portfolio.
• Foreign currency bonds are dominated by US dollar issuance, with the two categories sharing a correlation of r=0.994, while Japanese yen bonds show almost no correlation with CAD bonds (r=0.064), indicating a distinct issuance pattern.
• The distribution of bond values is heavily right-skewed, with a median of $187 million CAD versus a mean of $25,320 million CAD and a standard deviation of $135,144 million CAD, reflecting a small number of extremely large issuances.
• Government of Canada guaranteed bonds have declined over time as a share of total bonds outstanding, while direct bonds closely track and drive total portfolio growth.
• 38 statistical outliers (|z-score| > 2) were detected in total bonds outstanding, clustered specifically around July–November 2016, suggesting an unusual spike or structural reporting shift during that period.
• 1,319 rows containing negative values were identified exclusively within the 'Other currencies' category, likely reflecting accounting adjustments or reclassifications, with no negative values appearing in the 'Total currencies' aggregate.