Monday, September 28, 2009

Quarterly GDP 1920 to Present

Question

A PhD student is looking for quarterly GDP figures (nominal and real) from 1920 to the present. Such quarterly numbers are readily available from CANSIM from 1961 to 2008, but I could only find annual GDP for years prior to 1961. Historical Statistics of Canada cite National Income and Expenditure Accounts, Volume 1, The Annual Estimates, 1926-1974, (Catalogue 13-531) as their source but this publication, as its name implies, only presents annual data.

Are quarterly figures for 1920-1960 available through DLI or otherwise?



Answer

In the footnotes for the National Income and Expenditure and Related Aggregates, 1926 to 1976 (Series F1-152) section of "Historical Statistics of Canada", I found the following remark:

"The period of measurement for all the data given here is a year. Statistics Canada also publishes quarterly data which are available as far back as 1947."

In keeping with this, there is a publication called "National Income and Expenditure Accounts: Quarterly Estimates" (cat. no. 13-533) which provides gross national product and also gross domestic product at factor cost for 1947-1974.  If you don't have access to this publication, we could fax you Table 3, which shows both Gross National Product at market prices and Gross Domestic Product at factor cost.  However, be aware that due to revisions the data may not be comparable with the 1961+ data you have already found.



Follow-up question

We have that publication and I was able to verify that the quarterly data is presented therein. So as far as you can tell there is no quarterly data prior to 1947?



Follow-up answer

Yes that is correct.  I consulted with our historical catalogues and other reference materials (i.e. Guide to managing Statistics Canada publications in libraries, cat. no. 11-208E) and I could not find any quarterly data on this prior to 1947.

CMA/CA Designation for Communities in Newfoundland and Labrador

Question

I'm not sure if you would know the answers to some of these questions but perhaps you can direct me to someone who does.

1. I was looking at the tables for CMA NAICS 6.  For 1999- June 2007, the CMAs listed for Newfoundland and Labrador were: St. John's, Gander, Grand Falls/Windsor, Corner Brook, and Lab City.  In Dec. 2007, the CMAs changed.  They are now: St. John's, Grand Falls/ Windsor, Corner Brook, and Bay Roberts.  It seems we lost Gander and Lab City.  This isn't that unusual as the population size could have dropped below the minimum level for a CMA.  However, it seems odd that a town like Bay Roberts (also on the Avalon) would meet the criteria for a CMA.  Any ideas?

2. There are some NAICS codes I was looking for that appear from 1999-2001 and then they completely disappear (i.e. they are not even listed in the tables anymore).  Any ideas on why they would disappear?  Could they have been incorporated into another code?

3. There is one NAICS code (621910) listed in a document I received from a colleague.  This code is defined as Rescue Services.  However, it has never been listed in any of the CMA tables for 1999 to 2008.  You would think this would be an important code, especially here in NL, and it would definitely apply to our oceans sector research.  On the NAICS website (http://www.census.gov/cgi-bin/sssd/naics/naicsrch?code=621910&search=2002%20NAICS%20Search), if you type in this number, it comes up as Rescue Services and implies that it is a NAICS code for 1997, 2001, 2007, etc.  Any ideas as to why it is not included in the tables?

4. Do you know of a way to examine business births and deaths over time using the CBP data (besides going through each table year by year and recording numbers for each code)?  I wondered if there is a more streamlined and efficient way to uncover this info.



Answer



I have found the answers to several of the questions below:

1.  Bay Roberts is a Census Agglomeration (CA), not a CMA.  Among other criteria, CAs require an urban core population of only 10,000, whereas CMA status requires a total population of 100,000 (with at least 50,000 in the urban core).  Gander and Labrador City would also have been CA's before their populations dropped below 10,000. Here is the definition for both CAs and CMAs from the Census web site:

http://www12.statcan.ca/english/census06/reference/dictionary/geo009.cfm

2. The missing NAICS codes post-2001 may have to do with the 2002 revision of NAICS.  For a concordance between the 1997 and 2002 NAICS please consult the following link:

http://www.statcan.gc.ca/subjects-sujets/standard-norme/concordances/naics-scian97-02-eng.htm

3.  The NAICS link you provide below is to the U.S. version of NAICS 2002.  While NAICS is mostly comparable among the 3 North American countries, there are some exceptions, particularly at the 6-digit level.  For more information, please consult the following:

http://www.statcan.gc.ca/subjects-sujets/standard-norme/manual2002-manuel2002-eng.htm

4. Here is the response from the Division regarding your 4th question on business births and deaths:

"Unfortunately we can't produce the number of births and deaths in the CBP. The CBP contains the net changes to the establishment count (deaths and births). The CBP basically provides a snapshot of our business register at a specific time. The client should also be aware that the CBP is based on establishment counts and not counts of businesses."

There are some analytical studies on the Statistics Canada website that deal with entries and exits:

http://cansim2.statcan.gc.ca/cgi-win/cnsmcgi.pgm?Lang=E&SP_Action=Sub&SP_ID=2244&SP_Mode=2

Also, Section 2.1 of the following analytical study ("A Portrait of Entrants and Exits", Cat. no. 11F0019MPE No. 121) may be of interest to your client:

http://www.statcan.gc.ca/pub/11f0019m/11f0019m1999121-eng.pdf