Search

Many important economic variables are subject to revision. This article explains how, when, and why such revisions occur; how revisions to Canadian gross domestic product (GDP) compare with GDP revisions in some other countries; which GDP components are subject to the largest revisions; and how data revisions can affect policy decisions. The author finds that revisions to Canadian GDP tend to be smaller, on average, than those of some other countries, and that among the GDP components, exports and imports are most heavily revised.

We show that a large share of low-frequency (quarterly) movements in Canadian government bond yields can be explained by macroeconomic news, even though high-frequency (daily) changes are driven by other shocks. Furthermore, we show that US macro news—not domestic news— explains most of the quarterly variation in Canadian bond yields.

How much capital do banks need as a buffer to absorb severe shocks? By using historical stock market data, market-based stress tests help estimate the magnitude of capital buffers necessary to absorb severe but plausible shocks.

Similar to those of other forecasters, the Bank of Canada’s forecasts of global GDP growth have shown persistent negative errors over the past five years. This is in contrast to the pre-crisis period, when errors were consistently positive as global GDP surprised to the upside. All major regions have contributed to the forecast errors observed since 2011, although the United States has been the most persistent source of notable errors.

Our layered machine learning framework can enhance real-time transaction monitoring in high-value payment systems, which are a central piece of a country’s financial infrastructure. When tested on data from Canadian payment systems, it demonstrated potential for accurately identifying anomalous transactions. This framework could help improve cyber and operational resilience of payment systems.

This paper studies short-term forecasting of Canadian real GDP and its expenditure components using combinations of nowcasts from different models. Starting with a medium-sized data set, we use a suite of common nowcasting tools for quarterly real GDP and its expenditure components.

Using data from patents, citations, inter-sectoral sales and customs, we examine the international diffusion of technology through imports of sectoral knowledge and production inputs. We develop an instrumental variable strategy to identify the causal effects of technology embodied in imports on innovation and diffusion outcomes.

The break-even inflation rate (BEIR) is calculated by comparing the yields on conventional and Real Return Bonds. Defined as the average rate of inflation that equates the expected returns on these two bonds, the BEIR has the potential to contain useful information about long-run inflation expectations. Yet the BEIR has been higher, on average, and more variable than survey measures of inflation expectations, which may be explained by the effects of premiums and distortions embedded in the BEIR. Because of the difficulty in accounting for these distortions, the BEIR should not be given a large weight as a measure of long-run inflation expectations at this time. However, as the Real Return Bond market continues to develop, the BEIR should become a more useful indicator of inflation expectations. At present, it demonstrates no clear advantage over survey measures and even past inflation rates in forecasting near-term inflation.

Hurricanes disrupt oil production in the Gulf of Mexico because producers shut in oil platforms to safeguard lives and prevent damage. We examine the effects of these temporary oil supply shocks on real economic activity in the United States.

This paper considers the problem of estimating a linear model between two heavy-tailed variables if the explanatory variable has an extremely low (or high) value. We propose an estimator for the model coefficient by exploiting the tail dependence between the two variables and prove its asymptotic properties.