Business Entry and Exit Rates in Canada: A 30-year Perspective
by Ryan Macdonald, Economic Analysis Division
Start of text box
This article in the Economic Insights series describes the results of a data linkage project that created experimental long-term estimates of firm entry and exit rates for the Canadian business sector. It is part of a series of papers that examines firm dynamics using micro-economic data.
End of text box
Firm entry and exit are an important source of dynamism that is essential to a well functioning economy. The entry of new firms is an important source of productivity growth and technology adoption while exit removes less productive firms. Based on a new set of linked, experimental data for the period 1983/1984Note 1 to 2011/2012, Canada’s business-sectorNote 2 entry rate declined from 24.5% to 13.1%; and, the exit rate fell from 16.5% to 11.6% (Chart 1).
Entrants reflect an important aspect of economic dynamism because entry can be viewed as a form of experimentation which introduces new ideas, business models and technologies into the marketplace. Similarly, exits can be viewed as the end of unsuccessful experiments, which were selected out of the marketplace through competitive pressures. Measures of entry and exit are, therefore, a means of examining firm turnover, which is a major feature of micro-economic business dynamics.
Recent studies that focus on the years after 2000 offer mixed evidence that firm entry and exit rates declined in Canada. Ciobanu and Wang (2012) found that, based on the number of firms, a clear trend was not apparent in business-sector entry and exit rates between 2000 and 2008. Baldwin, Liu and Wang (2013) reported stable entry and exit rates in British Columbia, Alberta, Saskatchewan and Ontario, but negative trends in Atlantic Canada, Quebec and Manitoba. Criscuolo, Gal, and Menon (2014) found that, during the 2000s, the share of start-ups (firms less than three years old) declined in many nations, including Canada.Note 3
The entry rates and exit rates reported here are derived by calculating the number of entrants or exits divided by the population of active firms. This approach accounts for differences in the number of firms through time. These results extend the analyses in the previous work by showing that a downward trend in entry and exit rates is evident if the time series are lengthened with the addition of historical estimates. When post-2000 data are combined with historical values, the downward trend becomes clearer. These results are not inconsistent with the mixed conclusions in earlier research because the 2000-to-2010 period exhibits a less pronounced downward trend than do earlier periods.
The data show a long-run decline in the business-sector entry rate that was more than double the decline in the exit rate. Most of this difference was attributable to the 1980s when entry rates declined but exit rates increased. The entry rate declined by 5.7 percentage points between 1983/1984 and 1990/1991 while the exit rate increased by 2.6 percentage points. As a consequence, in the mid-1980s, the entry rate was 8.0 to 4.8 percentage points above the exit rate. By the 1990-1991 recession, the exit rate had risen to match the entry rate (which was trending downward) and the two then maintained a similar level until 1995/1996. In 1995/1996, the exit rate declined relative to the entry rate, and a more stable difference between entry and exit rates was established. The difference averaged 1.98 percentage points between 1996/1997, and the end of the sample period in 2011/2012.
Patterns similar across industries
From 1983/1984 to 2011/2012, entry rates in many industries declined by 7 to 11 percentage points, and exit rates tended to decline by 2 to 7 percentage points (Chart 2). This suggests similarity in the underlying forces that affect entry and exit rates—for instance, a general decline in experimentation—rather than idiosyncratic effects in particular industries.
Despite similarity in the magnitude of the declines in entry and exit rates, some changes were industry-specific as not all industries evolved in the same way or at the same pace. One example is manufacturing.
Manufacturing entry and exit rates exhibit many of the same features as overall business-sector entry and exit rates (Chart 3): a wider gap in the 1980s that was reduced by a rise in the exit rate; a long downturn in the entry rate; and long-term similarity in entry- and exit-rate trends.
However, after the late 1990s, unlike the business sector overall, in which the number of firms increased (the entry rate surpassed the exit rate), in the manufacturing sector, the entry rate fell below the exit rate. That is, the number of manufacturing firms began to decline. This coincides with the period when manufacturing’s share of value-added in the Canadian economy began to decline. By contrast, in the 1980s and 1990s, when the share of manufacturing in value-added rose, the entry rate surpassed the exit rate resulting in an increase in the number of manufacturing firms.
The differences in manufacturing entry and exit rates across decades reflect differences in the economic environment manufacturers faced. In the 1980s and 1990s, the Canadian dollar depreciated relative to the American dollar, and in the 1990s, the implementation of the Canada-United States Free Trade Agreement (FTA) and North American Free Trade Agreement (NAFTA) changed the composition of Canada’s manufacturing industry.Note 4 During these decades, the number of firms increased, and the share of manufacturing in value added in Canada moved counter to the international trend that saw a decline in the importance of this sector.Note 5
The implementation of the FTA and NAFTA, in particular, coincided with changes in firm entry rates for manufacturing. After the 1990-1991 recession, the entry rate rose from around 12.2% to 15.8% in 1996/1997, and then subsequently resumed a downward trend.
Is the economy becoming less dynamic?
Firm entry and exit rates in Canada fell during the last three decades. The decline in the entry rate was about twice as large as the decline in the exit rate, a pattern that was generally similar across industries. Experimentation and technology adaption through entry, and the process through which market forces lead unsuccessful firms to exit, weakened through time, and this is suggestive of a reduction in one aspect used to characterize the extent of economic dynamism in an economy. However, changes in entry and exit rates do not mean that Canada’s economy has become less dynamic. Entry and exit are only one aspect of the process of economic re-alignment that comprises economic dynamism. Other measures, such as firm turnover, changes in market share and employment re-allocation, are not addressed here and require further evaluation, in greater depth, and along additional dimensions.
Based on experimental linked data, a decline in firm entry and exit rates over the last 30 years is apparent. The magnitude of the decline was similar across industries, suggesting the decline was secular rather than the result of a compositional shift with growth favouring industries with lower entry rates and exit rates. Nor was the decline the result of one area of the economy adjusting to specific events, and producing a dynamic that differed substantially from another area. Although industry-specific shocks created industry-specific adjustments, they were not large enough to negate the evidence of widespread declines in entry and exit rates.
Baldwin, J.R., H. Liu, and W. Wang. 2013. Firm Dynamics: Firm Entry and Exit in the Canadian Provinces. The Canadian Economy in Transition, no. 030. Statistics Canada Catalogue no. 11-622-M. Ottawa: Statistics Canada.
Baldwin, J. R., and R. Macdonald. 2009. The Canadian Manufacturing Sector: Adapting to Challenges. Economic Analysis Research Paper Series, no. 057. Statistics Canada Catalogue no. 11F0027M. Ottawa: Statistics Canada.
Criscuolo, C., P.N. Gal, and C. Menon. 2014. The Dynamics of Employment Growth: New Evidence from 18 Countries. OECD Science, Technology and Industry Policy Papers no. 14. Paris: OECD Publishing.
Ciobanu, O., and W. Wang. 2012. Firm Dynamics: Firm Entry and Firm Exit in Canada, 2000-2008. The Canadian Economy in Transition, no. 022. Statistics Canada Catalogue no. 11-622-M. Ottawa: Statistics Canada.
Decker, R., J. Haltiwnager, R.S. Jarmin and J. Miranda 2014. The Secular Decline in Business Dynamism in the U.S. University of Maryland. Manuscript.
Hathaway, I., and R.E. Litan. 2014. Declining Business Dynamism in the United States: A Look at States and Metros. The Brookings Institution. Economic Studies at Brookings, May 2014.
Pilat, D., A. Cimper, K. Olsen, and C. Webb. 2006. The Changing Nature of Manufacturing in OECD Economies. DSTI Working Paper no. 9. Paris: Organization of Economic Co-operationand Development.
Rollin, A.M. 2012. Firm Dynamics: Employment Dynamics Arising from Firm Growth and Contraction in Canada, 2001 to 2009. The Canadian Economy in Transition, no. 024. Statistics Canada Catalogue no. 11-622-M. Ottawa: Statistics Canada.
United States Census Bureau. Business Dynamics Statistics. Last updated November 21, 2013. Available at:http://www.census.gov/ces/dataproducts/bds/ (accessed August 5, 2014).
Appendix: Linking the data
Statistics Canada’s Longitudinal Employment Analysis Program (LEAP) tracks firm entry and exit. The program follows firms through time by using tax files, particularly administrative data on T4 filings and the associated business numbers. By linking administrative files for multiple years, the lifespan of a particular firm can be tracked.
A new LEAP vintage is constructed each year. Through implementing rules for mergers, acquisitions and divestitures, LEAP creates a particular firm structure unique to that vintage. As a result, with each year that is added to the dataset, the number of firms changes. These changes have a greater effect on the number of firms than the growth rate of firms. Consequently, the long-run estimates of firm entry and exit constructed here use growth rates from past vintages to back-cast current year estimates.
To form industry estimates for firm entry and exit, it is necessary to transform data from past vintages from the Standard Industrial Classification (SIC) to the North American Industry Classification System (NAICS). Based on information from the LEAP files for overlapping years, a concordance between SIC and NAICS was created. This concordance was used to transform the SIC codes from SIC-based vintages into NAICS industry codes.
For a number of industries, correlations between NAICS-based data and the SIC-to-NAICS converted data were lower than desired. Because the NAICS 61 (education) and NAICS 62 (health care and social assistance) industry links performed poorly, these industries are not reported, and the business-sector aggregate omits them. NAICS 52 (finance and insurance), NAICS 53 (real estate) and NAICS 55 (management of companies) also yielded values that were not ideal for linking, and so were grouped into finance, insurance, real estate and leasing (FIREL), which provided more appropriate results. The aggregate value, rather than the individual NAICS industries, is reported. Finally, the NAICS 22 (utilities) industry contained a number of aberrant observations; its values were deemed to be of insufficient quality to publish and report separately.
To back-cast the data, historical estimates are created at two levels of aggregation. First, the top-level business sector is back-cast. The business sector is defined as all industries excluding NAICS 61 (education), NAICS 62 (health care and social assistance) and NAICS 91 (public administration). Because firms’ industry allocation can change over time, the business sector offers the most stable link.
Next, a back-cast is performed for each industry. These industry-based estimates are used to calculate a bottom-up business sector estimate. This estimate is not identical to the top-level value, which has the strongest link. To produce industry estimates that match the top-level business sector estimate, each industry’s share, in each year, of the bottom-up business sector estimate is calculated, and these shares are used to distribute the top-level link across industries. In this way, the strongest aggregate link is combined with information from the industry links so as to ensure aggregation across industries to the top-level business-sector data.
The results for the entry and exit rate by industry are presented in Table 1 and Table 2.
- Date modified: