COMPETITIVE ADVANTAGE ANALYSIS

Transcription

1 What Makes Saskatchewan Tick? COMPETITIVE ADVANTAGE ANALYSIS Detailed Statistical Report September 2005

2 WHAT MAKES SASKATCHEWAN TICK? COMPETITIVE ADVANTAGE ANALYSIS September 2005 Sask Trends Monitor th Avenue Regina, Saskatchewan S4N 1H1 Tel: Internet:

3 ii

4 Table of Contents SECTION 1 INTRODUCTION AND BACKGROUND... 1 SECTION 2 LOCATION QUOTIENTS Using Employment as an Indicator Using GDP as an Indicator Summary SECTION 3 SHIFT SHARE ANALYSIS Using Employment as an Indicator Using GDP as an Indicator Summary SECTION 4 LEADING/LAGGING ANALYSIS Using Employment as an Indicator Using GDP as an Indicator Summary SECTION 5 INDUSTRY TARGETING Using Employment as an Indicator Using GDP as an Indicator Summary SECTION 6 SUMMARY iii

5 iv

6 List of Tables Table 1.1 Competitive Advantage Analysis... 8 Table 2.1 Interpretation of Location Quotients... 9 Table 2.2 Employment Based Location Quotients for Saskatchewan Relative to Western Provinces Table 2.3 Employment Based Location Quotients for Saskatchewan Relative to Midwestern States and Prairie Provinces, Paid Workers Only Table 2.4 GDP Based Location Quotients for Saskatchewan Relative to Western Provinces Table 2.5 GDP Based Location Quotients for Saskatchewan Relative to Midwestern States and Prairie Provinces, Table 2.6 Location Quotients for Saskatchewan Table 3.1 Shift Share Components for Employment Growth from 1992 to 2004, Comparison with Western Canada Table 3.2 Shift Share Components for Paid Employment Growth from 2001 to 2003, Comparison with Midwest States and Prairie Provinces Table 3.3 Shift Share Components for Real GDP from 1997 to 2004, Comparison with Four Western Provinces. 30 Table 3.4 Shift Share Components for Real GDP from 1998 to 2002, Comparison with Midwest States and Prairie Provinces Table 3.5 Provincial Effect in the Shift/Share Analysis for Saskatchewan Table 4.1 Carvalho Classification System Table 4.2 Leading/Lagging Classification Relative to Four Western Provinces, Employment Measure, 1992 to Table 4.3 Leading/Lagging Classification Relative to Midwest States and Prairie Provinces, Paid Employment Measure, 2001 to Table 4.4 Leading/Lagging Classification Relative to Four Western Provinces, Real GDP Measure, 1997 to Table 4.5 Leading/Lagging Classification Relative to Midwest States and Prairie Provinces, Real GDP Measure, 1998 to Table 4.6 Leading/Lagging Classification for Saskatchewan Table 5.1 Industry Targeting Classification System Table 5.2 Industry Targeting Classification, Employment in Saskatchewan Relative to Four Western Provinces, 1992 to Table 5.3 Industry Targeting Classification, Paid Employment in Saskatchewan Relative to Midwest States and Prairie Provinces, 2001 to Table 5.4 Industry Targeting Classification, GDP in Saskatchewan Relative to Four Western Provinces, 1997 to Table 5.5 Industry Targeting Classification, GDP in Saskatchewan Relative to Midwest States and Prairie Provinces, 1998 to Table 5.6 Industry Targeting Classification for Saskatchewan Relative to the Four Western Provinces Table 5.7 Industry Targeting Classification for Saskatchewan Relative to the Midwest States and Prairie Provinces Table 6.1 Comparative Advantage Summary for Agriculture Table 6.2 Comparative Advantage Summary for Mining and Oil/Gas Extraction Table 6.3 Comparative Advantage Summary for Utilities Table 6.4 Comparative Advantage Summary for Construction Table 6.5 Comparative Advantage Summary for Agriculture Table 6.6 Comparative Advantage Summary for Accommodation and Food Services Table 6.7 Comparative Advantage Summary for Professional, Scientific, and Technical Services List of Figures Figure 1.1 Region Chosen for Midwest Provinces and States, (showing relative sizes of state and provincial economies)... 3 Figure 1.2 Real Gross Domestic Product (chained 1997 dollars), Western Provinces, 1991= Figure 1.3 Employment in the Western Provinces, 1991= Figure 2.1 Employment Based Location Quotients, Saskatchewan Compared with the Four Western Provinces, 1992 and Figure 2.2 Location Quotients for Paid Employment, Saskatchewan Compared with the Midwest States and Prairie Provinces, v

7 Figure 2.3 GDP Based Location Quotients, Saskatchewan Compared with the Four Western Provinces, 1997 and Figure 2.4 GDP Based Location Quotients, Saskatchewan Relative to the Midwest States and Prairie Provinces, Figure 3.1 Provincial Effect Using Employment, Saskatchewan Compared with the Four Western Provinces, 1992 to Figure 3.2 The Provincial Effect Using Paid Employment, Saskatchewan Compared with the Midwest States and Prairie Provinces, 2001 to Figure 3.3 The Provincial Effect Using Real GDP, Saskatchewan Compared with the Four Western Provinces, 1997 to Figure 3.4 The Provincial Effect Using Real GDP, Saskatchewan Compared with the Midwest States and Prairie Provinces, 1998 to Figure 4.1 Leading/Lagging Classification using Employment, Saskatchewan Relative to the Four Western Provinces, 1992 to Figure 4.2 Leading/Lagging Classification using Paid Employment, Saskatchewan Relative to the Midwest States and Prairie Provinces, 2001 to Figure 4.3 Leading/Lagging Classification Relative to Four Western Provinces, GDP Measure, 1997 to Figure 4.4 Leading/Lagging Classification, Saskatchewan Relative to the Midwest States and Prairie Provinces, GDP Measure, 1998 to Figure 5.1 Industry Targeting using Employment, Saskatchewan Compared with the Four Western Provinces, 1992 to Figure 5.2 Industry Classification Using Paid Employment Growth, 2001 to 2003, Relative to the Midwest States and Prairie Provinces Figure 5.3 Industry Classification Using Growth in Real GDP, 1997 to 2004, Saskatchewan Compared with the Four Western Provinces Figure 5.4 Industry Classification Using Growth in Real GDP, 1998 to 2002, Saskatchewan Compared with the Midwest States and Prairie Provinces vi

8 SECTION 1 INTRODUCTION AND BACKGROUND Governments, businesses, individuals, and organizations involved in community and economic development are always searching for a way to forecast patterns of economic growth. They do this for all the right reasons, of course, wanting to encourage economic growth in areas that have the most potential to be successful and avoid devoting scarce resources to economic areas and opportunities that are more likely to fail. Unfortunately, forecasting economic trends is extremely difficult. The market has a tendency to move in unpredictable ways because of new technology (think of the Internet), changes in politics or world events (think of September 11, 2001), or other factors (think of BSE). This document is not a forecast of future opportunities. It is, instead, an examination of past trends with a view to identifying industry sectors that have under-performed or over-performed relative to other industries and other regions. Perhaps this will enable an informed debate about where the province s economy is headed and about what makes Saskatchewan tick. This report summarizes the results of a competitive advantage analysis 1 for the Saskatchewan economy. A competitive advantage analysis looks at recent industry trends in comparison with trends in other economies and other industries. The research was conducted by Doug Elliott, the principal of QED Information Systems Inc. and publisher of Sask Trends Monitor. The opinions expressed in this document do not necessarily represent those of the Chamber of Commerce, the Government of Canada, or the members of the research committee which was charged with overseeing the research. Responsibility for the accuracy of the data and the validity of the conclusions reached remains with the author. Choosing the Economic Measures In theory a competitive advantage analysis can be done for any kind of economic measure. In practise, there are only two kinds of economic statistics employment and the gross domestic product that are comprehensive, consistently available over time for different regions and industry groups. There are advantages and disadvantages to using each of these measures. Using employment as a measure of economic activity has the advantage of being easy to understand. A job is a readily identifiable measure of economic success for an individual if not an economy. The data from Statistics Canada s Labour Force Survey (LFS) are both up-to-date and reliable and historically comparable information by industry is available back to the mid 1980s. But there are difficulties with employment as a measure of activity at the industry level. The first is that changes in productivity can affect changes in employment; a firm or industry may be more successful by every other measure except employment. If an individual firm or an industry group has revenue growth, higher salaries, and profit increases with the same number of employees, a measure of employment will not indicate success. 1 The model for Competitive Advantage Analysis used in this report is based on research done for the Ontario Ministry of Municipal Affairs by economist Dr. Emanuel Carvalho from the University of Waterloo. Competitive Advantage Analysis September 2005 Page 1

9 A second difficulty is related to a uniquely Saskatchewan phenomenon, namely the presence of multiple job holders, people who simultaneously hold two or more jobs. In 2004, for example, 8% of persons employed in the province were multiple job holders, the highest rate in Canada. The LFS measures the characteristics, including the industry, of only the main job the one at which the respondent works the most hours. We undoubtedly misrepresent agricultural employment in the province, for example, when we measure only employment for the main job. A third difficulty is with part-time employment specifically and hours of work generally. Replacing two part-time employees with a single full-time employee working the same aggregate hours will look, in the statistics, like an increase in employment. Increased hours for either part-time or fulltime employees will not register. A more appropriate measure might be hours worked rather than employment but in practise this would make little difference because changes in hours of work are relatively slow to occur and tend to happen in Saskatchewan at the same rate as in other provinces. The Gross Domestic Product (GDP) is a more comprehensive measure of economic activity. The GDP is a measure of the value of goods and produced by an economy and is the statistic used most often by economists. Adjusted for changes in price levels, real GDP is a measure of the value added by an economy. Besides being harder for a non-economist to understand, the difficulty with GDP measures either real or nominal is accuracy and availability. At the provincial level, calculation of GDP requires a good deal of detailed raw data about international and interprovincial trade, corporate profits, wage rates, and the volume and value of goods and produced. Much of this hard data is not available so provincial GDP figures are based on econometric models to a large extent and prone to error and periodic retroactive revisions that can be significant. To compound the difficulty with gross domestic product, Statistics Canada has recently begun a process to re-base provincial GDP values to a new industry classification scheme - the North American Industry Classification System (NAICS). Data for real GDP by province is only available since Choosing the Comparison and the Time Frame A competitive advantage analysis requires: i) another economy against which the Saskatchewan economy can be compared; and ii) a time frame over which the comparison is made. The choices made for these parameters are important. Saskatchewan s resource-based economy has experienced a number of booms and busts over the years and the choice of the starting and ending points will have an impact on how well or how poorly the economy performs over a period of time. The economy chosen for the comparison should reflect the same kinds of economic opportunities as those available in Saskatchewan. The obvious choice for a comparison economy is Canada as a whole. This has some risks, however, because the national economy is not at all like the Saskatchewan economy. Dominated by the manufacturing sectors in populous Ontario and Quebec, any comparison with the national economy would certainly make Saskatchewan s resource-based economy look different. A focus on the prairie provinces, on the other hand, may be too narrow. Saskatchewan accounts for one sixth of employment or GDP in the prairie provinces and one would normally like to find a comparable region in which we play a smaller role. An analysis that expanded the prairies to Competitive Advantage Analysis September 2005 Page 2

10 include the northern states would also be interesting because this kind of comparison would include economies under a different political régime but with similar climates and natural resources. Another alternative is to include British Columbia which also has a significant but different kind of resource base. In the end, the research committee decided that two comparisons would be made; i) one between Saskatchewan and the four western provinces, and ii) one between Saskatchewan and the midwest states combined with the prairie provinces. Figure 1.1 shows the states and provinces chosen for the latter comparison. Choosing a time frame is also important. We are measuring changes over time so both the starting point and the end point is important. The end point would be the most recent year available ideally 2004 but the starting point could be any time after 1986 when comparable data were produced. Obviously an earlier start date means that changes will be measured over a longer period and less subject to annual fluctuations. Figure 1.1 Region Chosen for Midwest Provinces and States, (showing relative sizes of state and provincial economies) 2002 GDP measured in 1997$ Alberta C$120 B Saskatchewan C$28 B Manitoba C$32 B Montana US$23 B North Dakota US$19 B Minnesota US$193 B Wyoming US$20 B South Dakota US$24 B Nebraska US$58 B Iowa US$94 B 2002 GDP measured in 2000$ Kansas US$86 B Missouri US$179 B Competitive Advantage Analysis September 2005 Page 3

11 The last national recession (defined as a decline in real GDP) occurred in The Manitoba economy was also in recession that year whereas the B.C. and Alberta economies slowed but did not decline. Saskatchewan s economy went into recession one year later, in 1992 (see Figure 1.2). As well, 1992 marks a point where employment in both the national economy and in three of the four western provinces reached a low (see Figure 1.3) so seems a good starting point for the analysis. In effect, we will be measuring, for the comparison with the four western provinces, changes over the past twelve years from 1992 to 2004 in the analysis of employment that follows. The comparison with the midwest states and prairie provinces requires the use of a much shorter time frame and a slightly different measure of employment. Data on employment in the USA is published by the Bureau of Labour Statistics (BLS) and is derived from payroll records rather than a telephone survey. The information therefore excludes the self-employed because there are no payroll records to measure employment. Accordingly, the comparison with the midwest states and prairie provinces is made for only paid workers, effectively minimizing agriculture as an industry because, for example, only one in six persons employed in Saskatchewan agriculture is a paid worker. Other industries with a large proportion of the self-employed such as real estate, professional and technical, retail trade, and accommodation and food are also affected. Other problems arise with the comparison because the BLS has only been publishing data using the same industry groups as Statistics Canada since 2001 so there is a lack of historical data from the USA. For real GDP, the time frame is limited by the availability of the data the seven years from 1997 to 2004 were chosen for the comparison with the four western provinces. For the midwest states and prairie provinces, real GDP is available only from 1998 to Figure 1.2 Real Gross Domestic Product (chained 1997 dollars), Western Provinces, 1991= = Alberta B.C. Manitoba Saskatchewan Competitive Advantage Analysis September 2005 Page 4

12 Choosing the Industry Sectors The choice of industry sectors to be used in the analysis was determined by availability. The Labour Force Survey conducted monthly by Statistics Canada is the only source of current employment information so, by necessity, it was used as the main data source for the analysis of employment. Statistics Canada uses the NAICS (North American Industry Classification System) to classify industries in the LFS. In practical terms, the analysis is limited to the detail available in the NAICS classification scheme and published in the LFS. The only change was to combine two smaller industry groups to ensure reliability of the data (forestry and logging was combined with fishing, hunting, and trapping). This results in nineteen different industry groups for the comparison of employment in western Canada. For GDP, the industry groups are published with slightly different groupings resulting in eighteen industry groups. The employment data from the LFS is based on annual averages for the years in question, that is, the average of twelve monthly snapshots of employment 2. To be employed, a person must have done work for pay or profit. Employed persons can be paid workers, self-employed, or the so-called unpaid family workers, namely those who work without pay in a family farm, business or professional practice. Those who are absent from their job or business because of a vacation, illness, labour dispute, or other reason are still considered as employed. Persons with two or more simultaneous jobs are counted only once, classified in their main job. Figure 1.3 Employment in the Western Provinces, 1991= = Alberta B.C. Manitoba Saskatchewan The LFS does not survey residents on Reserve so the employment figures for Saskatchewan will be somewhat understated, particularly in the public sector, compared with those in other provinces. Competitive Advantage Analysis September 2005 Page 5

13 The Bureau of Economic Analysis (BEA) in the US is responsible for publishing the GDP and the BEA uses a different approach to the public sector in the economic accounts. In effect all government related activities, what we call the public sector, is classified as government. So for example, the public sector part of education is classified as part of government whereas in Canada, it is considered part of the education sector. This affects the comparison between Saskatchewan and the midwest states and prairie provinces group in any sector where there is a significant government presence in either the USA or Canada. Besides public administration, these include transportation (the postal service), health care, education, and utilities. A more complete description of the industry categories used in this analysis follows. Agriculture Forestry, fishing, hunting Mining, oil and gas extraction Utilities Construction Manufacturing This category includes grain and cattle farming, market gardens, intensive livestock operations, and related to agriculture such as crop spraying, custom harvesters, seed cleaning, and animal breeding. Logging and forestry, commercial and sustenance fishing and trapping are included in this category. Note that the logging industry only measures activity in the forest (including silviculture), and doesn t include processing activities. This category includes oil and gas exploration and production, above and below ground mining, and contract drilling for oil and natural gas. Oil refineries are classified as manufacturing establishments. Establishments engaged in the production and distribution of electricity or the distribution of natural gas and water are classified as utilities. (Telephone companies are considered as part of the information sector.) In Saskatchewan, this category will be dominated by SaskEnergy and SaskPower. This sector comprises establishments primarily engaged in constructing, repairing and renovating buildings and engineering works, and in subdividing and developing land. These establishments may operate on their own account or under contract to other establishments. They may produce complete projects or subcontract work to other establishments. Manufacturing establishments take raw materials and transform them into finished products, in the sense that they are ready to be used or consumed, or semi-finished products, in the sense of becoming a raw material for an establishment to use in further manufacturing. Manufacturing establishments may own the materials which they transform or they may transform materials owned by other establishments. In Saskatchewan, this category includes traditional manufacturing activities as well as food processing (meat processing plants and grain milling). Oil refineries are also considered as manufacturing establishments. Non-durable goods include food, beverages, clothing, textiles, chemicals, paper products, plastics and rubber products. Durable goods include machinery and equipment, steel and metal products, transportation equipment, wood products, furniture and electrical/electronic equipment. Wholesale trade Retail trade Transportation and warehousing The wholesaling process is generally an intermediate step in the distribution of merchandise; many wholesalers are therefore organized to sell merchandise in large quantities to retailers, businesses, and institutional customers. Some wholesalers, in particular those that supply non-consumer capital goods, sell merchandise in single units to final users. Farm machinery dealers and lumber yards are considered as wholesale establishments. The retail category includes establishments that sell merchandise (not ) to the general public. Non-store retailers such as vending machine operators are included as well. Establishments in this category transport people or goods by rail, air, or road. Included are trucking companies, the post office, courier, pipelines, grain terminals, airlines, taxis, and scheduled, charter, or school bus. Competitive Advantage Analysis September 2005 Page 6

14 Finance and insurance Real estate and leasing Professional, scientific, and technical Business building and other support Education service Health and social assistance Information, culture, and recreation Accommodation and food Personal and household Public administration This category includes both deposit-taking financial institutions such as banks and credit unions as well as financial management firms. Insurance companies and insurance agencies are also included. This category is dominated by self-employed real estate agents although it also includes establishments that rent or lease equipment or vehicles. These kinds of establishments provide primarily to other businesses. The category includes, for example, management and computer consulting firms, law offices, advertising agencies, accounting, architectural, and engineering. This category includes head offices and companies that provide waste management or cleaning to buildings. It also includes other to business such as temporary employment agencies, investigation and security, and office administration. Education include elementary and secondary education, post-secondary institutions and private vocational schools. Health and social includes establishments that are involved in the delivery of health such as hospitals, ambulance, physicians or dentists offices or social such as special care homes, child care, or for those with a disability. This category includes the media, telecommunications, performing arts companies, amusement parks, museums, casinos, sports and cultural venues as well as agents for artists and athletes. These are establishments primarily engaged in providing accommodation such as hotels, motels, vacation farms, and campgrounds or food and beverage such as lounges, restaurants, and catering. This category is dominated by companies that generally provide personal or household to individuals. It includes hair salons, funeral homes, and laundry as well as lawn care companies, and auto repair shops. Membership organizations such as employee organizations, churches, trade unions, and social advocacy organizations are also included. Public administration includes all levels of government federal, provincial, town, municipal, and First Nation and the agencies which operate at arm s length from government such as Workers Compensation and the Human Rights Commission. It does not include government enterprises such as crown corporations. Note that the public sector industry groups public administration, health care and education, for example are included in the analysis for completeness. There is no suggestion that these are economic drivers for the provincial or regional economies. Competitive Advantage Analysis September 2005 Page 7

15 Table 1.1 Competitive Advantage Analysis Comparison group Employment Economic indicator Real Gross Domestic Product Four western provinces Midwest states and prairie provinces 19 industry groups 1992 to industry groups 2001 to industry groups 1997 to industry groups 1998 to 2002 Table 1.1 summarizes the time frames and industry groups used in this competitive advantage analysis. Competitive Advantage Analysis September 2005 Page 8

16 SECTION 2 LOCATION QUOTIENTS A location quotient identifies the level of specialization in a geographic region. In simple terms it measures the concentration of certain industry sectors in the region relative to a reference economy either the four western provinces or the midwest states and prairie provinces. In mathematical terms, the location quotient for each individual sector (using employment data comparison to the four western provinces, for example) is where: LQ SECT = the location quotient for an industry sector SASKEMPL SECT = employment in the industry sector in Saskatchewan SASKEMPL = total employment in Saskatchewan WCEMPL SECT = employment in the industry sector in western Canada WCEMPL = total employment in western Canada The analysis for the midwestern states and provinces substitutes the aggregate of the provinces and states for the four western provinces. The analysis for GDP uses real GDP figures in the equation instead of employment figures. Table 2.1 shows the range of possible location quotients arising from the formula and their suggested interpretation. The interpretation of location quotients is not particularly complex; we are simply measuring employment or economic concentration in the province. Industry groups that dominate in the province will have higher location quotients and ones that are relatively scarce will have lower location quotients. Not that a low (or high) location quotient doesn t necessarily mean the industry group is unimportant (or important) in Saskatchewan, just that it is less so (or more so) than in the comparison economy. Table 2.1 Interpretation of Location Quotients Location Quotient Label Interpretation less than 0.75 Low Provincial needs are not being met by the sector. The province is probably importing goods and in this sector to 1.24 Medium Most local needs are being met by the sector. The province will probably be both importing and exporting goods and in this sector to 4.99 High The sector is serving needs beyond the border, exporting goods and in this sector to other provinces. More than 5.00 Very High This is indicative of a very high level of local dependence on the sector, typically in a single-industry community. Competitive Advantage Analysis September 2005 Page 9

17 2.1 Using Employment as an Indicator Relative to the four western provinces, all but three of the nineteen industry groups have a location quotient for employment that is classified as either low or medium in 2004 (see Table 2.2 and Figure 2.1). The three industry groups with a high location quotient are agriculture, the resource sector, and utilities. Note that an increase in the location quotient can arise if employment in Saskatchewan increases as a proportion of total employment or if it declines as a proportion of employment in the comparison group. In two of these three cases, the location quotient has increased from a medium category in 1992 to a high category in Although there are changes in the location quotient for the other seventeen industry groups, all are classified in 2004 in the same category as in The educational group is almost an exception the location quotient has increased from 1992 to 2004 to the extent that education is almost in the high category. The other observation that can be made about the comparison with the four western provinces is that Saskatchewan s location quotients tend to be increasing. In thirteen of the nineteen industry groups, the location quotient is higher in 2004 than in This is an indication that the province s economy is becoming more diversified employment in agriculture declined by 42% over the twelve years and there was an increase of 18% in non-agricultural employment. Several of the location quotients are low enough to suggest that the province is importing a large proportion of goods/ in these industry groups. These include the forestry/fishing/hunting group, the manufacturing sector, and the professional, scientific, and technical group. Table 2.2 Employment Based Location Quotients for Saskatchewan Relative to Western Provinces Category in 2004 Agriculture High Forestry, fishing, hunting Low Mining and oil and gas extraction High Utilities High Construction Low Manufacturing Low Wholesale trade Medium Retail trade Medium Transportation and warehousing Medium Finance and insurance Medium Real estate and leasing Low Professional, scientific and technical Low Business, building and other support Low Educational Medium Health care and social assistance Medium Information, culture and recreation Medium Accommodation and food Medium Personal and household Medium Public administration Medium Competitive Advantage Analysis September 2005 Page 10

18 Figure 2.1 Employment Based Location Quotients, Saskatchewan Compared with the Four Western Provinces, 1992 and 2004 Agriculture Utilities Mining and oil and gas extraction Educational Health care and social assistance Public administration Finance and insurance Wholesale trade Retail trade Personal and household Information, culture and recreation Accommodation and food Transportation and warehousing Real estate and leasing Business, building and other support Construction Manufacturing Professional, scientific and technical Forestry, fishing, hunting low medium high Competitive Advantage Analysis September 2005 Page 11

19 Midwestern States and Prairie Provinces As noted in Section 1, the data on employment in the USA is restricted to paid employment and available only since Table 2.3 shows the location quotient, based on paid workers, for Saskatchewan relative to the midwest states and prairie provinces. In the categories using paid employment, retail and wholesale trade joins agriculture, the resource sector, and utilities as having a high location quotient in Saskatchewan in The location quotient for the resource sector is much higher in this comparison than with the comparison to four western provinces. Two of the industries dominated by the public sector, namely education and government proper, remain high in this comparison but less so than in the comparison with Western Canada. The other public sector industry health care has a lower location quotient relative to the midwest states and prairie provinces than in does in the comparison with the four western provinces. The comparison with four western provinces showed manufacturing, construction, the professional, scientific and technical group, and the business, building, and other support group as having a low location quotient in Saskatchewan. The location quotients relative to the midwest states and prairie provinces are still low for these industries but construction and the professional group have moved up to a medium classification. Table 2.3 Employment Based Location Quotients for Saskatchewan Relative to Midwestern States and Prairie Provinces, Paid Workers Only 2003 Category Agriculture, forestry, fishing and hunting 2.3 High Mining, oil and gas 3.1 High Utilities 1.3 High Construction 0.8 Medium Manufacturing 0.5 Low Retail and wholesale trade 1.3 High Transportation and warehousing 1.1 Medium Finance, insurance, real estate and leasing 1.0 Medium Professional and technical 0.8 Medium Business, building and other support 0.4 Low Educational 1.1 Medium Health care and social assistance 1.0 Medium Information, arts, entertainment, and recreation 1.1 Medium Accommodation and food 0.9 Medium Personal and household 1.2 Medium Public Administration 1.2 Medium Competitive Advantage Analysis September 2005 Page 12

20 Figure 2.2 Location Quotients for Paid Employment, Saskatchewan Compared with the Midwest States and Prairie Provinces, 2002 Mining, oil and gas 3.1 Agriculture, forestry, fishing and hunting 2.3 Retail and wholesale trade Utilities Public Administration Personal and household Educational Information, arts, entertainment, and recreation Transportation and warehousing Finance, insurance, real estate and leasing Health care and social assistance Accommodation and food Construction Professional and technical Manufacturing Business, building and other support low medium high Competitive Advantage Analysis September 2005 Page 13

21 2.2 Using GDP as an Indicator Relative to the four western provinces, all but five of the eighteen industry groups have a location quotient for real GDP that is classified as medium in 2004 (see Table 2.4 and Figure 2.3). The two industry groups with a high location quotient are agriculture and the resource sector. Those that are considered as low are the professional, scientific, and technical group, the administrative and support, waste management group, and accommodation and food. The location quotients for all three have increased since These classifications are similar to the ones derived using employment rather than GDP. A comparison with the employment data, which is hampered by slightly different industry groupings, shows that utilities is ranked medium in terms of GDP but high in terms of employment. This would indicate that the utility sector in Saskatchewan has a typical level of output but relatively high levels of employment. Manufacturing and construction were both ranked low in terms of employment but medium in terms of GDP. This would indicate the opposite, namely a relatively low level of employment but relatively high output for manufacturing and construction. The largest declines over the seven years have been in construction, accommodation and food, and the professional, scientific, and technical group. The largest increases have been in transportation/warehousing and manufacturing. Table Category in 2004 Agriculture, forestry, fishing and hunting High Mining and oil and gas extraction High Utilities Medium Construction Medium Manufacturing Medium Wholesale trade Medium Retail trade Medium Transportation and warehousing Medium Information and cultural industries Medium Finance and insurance, real estate, renting/leasing, management of companies and enterprises GDP Based Location Quotients for Saskatchewan Relative to Western Provinces Medium Professional, scientific and technical Low Administrative and support, waste management and remediation Low Educational Medium Health care and social assistance Medium Arts, entertainment and recreation Medium Accommodation and food Low Other Medium Public administration Medium Competitive Advantage Analysis September 2005 Page 14

22 Figure 2.3 GDP Based Location Quotients, Saskatchewan Compared with the Four Western Provinces, 1997 and 2004 Agriculture, forestry, fishing and hunting Mining and oil and gas extraction Utilities Public administration Health care and social assistance Educational Wholesale trade Transportation and warehousing Personal and household Information and cultural industries Retail trade Finance, insurance, real estate, leasing, head offices Arts, entertainment and recreation Construction Manufacturing Accommodation and food Administrative and support, waste management and remediation Professional, scientific and technical low medium high Competitive Advantage Analysis September 2005 Page 15

23 Midwestern States and Prairie Provinces The location quotients for Saskatchewan relative the midwest states and prairie provinces using real GDP are shown in Table 2.5 and graphically in Figure 2.4 As noted in Section 1, there are comparison difficulties in the industry sectors dominated by the public sector in these data. Once again, however, agriculture and the resource sector are classified as high in terms of their location quotient. These two sectors are joined by three others that are affected by comparability issues utilities, transportation and warehousing, and educational. Two industry groups that had a relatively low location quotient in other comparisons are higher here construction and accommodation and food. The same two industry groups that were classified as low in terms of their location quotient in the other comparison, namely manufacturing and the professional, scientific, and technical group, remain low in this comparison. Government is also in this category but only because it is more broadly defined in the US data. Table 2.5 GDP Based Location Quotients for Saskatchewan Relative to Midwestern States and Prairie Provinces, Category Agriculture, forestry, fishing, and hunting 2.4 High Mining, oil and gas extraction 3.7 High Utilities* 1.3 High Construction 1.1 Medium Manufacturing 0.6 Low Wholesale trade 1.0 Medium Retail trade 0.8 Medium Transportation and warehousing* 1.6 High Information and culture 0.8 Medium Professional and technical 0.4 Low Educational * 3.6 High Health care and social assistance* 1.0 Medium Arts, entertainment, and recreation 0.8 Medium Accommodation and food 0.9 Medium Personal and household 1.1 Medium Other ** 0.8 Medium Government* 0.6 Low * comparability in these categories is compromised because the US midwestern states include all public sector activities in the government category ** finance, insurance, real estate, rental and leasing, head offices, waste management and remediation Competitive Advantage Analysis September 2005 Page 16

24 Figure 2.4 GDP Based Location Quotients, Saskatchewan Relative to the Midwest States and Prairie Provinces, 2002 Mining, oil and gas 3.7 Educational * 3.6 Agriculture, forestry, fishing, and hunting 2.4 Transportation and warehousing* Utilities* Construction Personal and household Health care and social assistance* Wholesale trade Accommodation and food Arts, entertainment, and recreation Other ** Information and culture Retail trade Manufacturing Government* Professional and technical low medium high * "Government" in the USA includes all public sector activities including health, education, and the postal service ** Finance, insurance, real estate, rental and leasing, head offices, waste management and remedial Competitive Advantage Analysis September 2005 Page 17

25 2.3 Summary Table 2.6 summarizes the location quotients for Saskatchewan from this section. A high location quotient (1.25 or more) is indicative of a concentration in that particular sector whereas a low location quotient (less than 0.75) means that the province will be importing the majority of goods or produced by this sector. Several industries have a high location quotient for Saskatchewan relative to both the four western provinces and the midwest states and prairie provinces. Agriculture and the resource sector both have consistently high location quotients regardless of the measure and the comparison. The resource sector is particular strong relative to the midwest states and prairie provinces. The utilities sector is classified as having a high location quotient in three of the four comparisons. The only exception is the GDP measure relative to western Canada. The public sector groups public administration, health care and social assistance, and education have a relatively high location quotient but they are generally classified as medium in the methodology. Several industries have a low location quotient for Saskatchewan relative to both the four western Canadian provinces and the midwest states and prairie provinces. Manufacturing and the Professional, Scientific, and Technical Services group has a low location quotient in all four comparisons. Construction typically has a low location quotient although the paid employment comparison with the midwest states and prairie provinces shows a high location quotient. This would suggest that Saskatchewan has a relatively low proportion of self-employed construction workers compared with the midwest states and prairie provinces. Although ranked as having a medium concentration by the methodology, the accommodation and food sector has a location quotient that is consistently well below 1.0. Several other industries are average in the sense that their location quotient is near 1.0 in the four comparisons. Retail and wholesale trade combined have a medium location quotient although wholesale trade is often a little higher than retail trade. This is undoubtedly because of the number of farm machinery dealers in the province. One exception is the paid employment measure for the midwest states and prairie provinces which would suggest that the province has relatively few self-employed individuals in this sector. Depending on the measure, the transportation and warehousing sector is near the reference point of 1.0 somewhat below for employment and somewhat above for the GDP measure. Personal and household are near the reference point of 1.0 in all four comparisons. Competitive Advantage Analysis September 2005 Page 18

26 Table 2.6 Location Quotients for Saskatchewan Relative to Western Canada Relative to Midwest States and Prairie Provinces Employment (2004) GDP (2004) Employment (2003) GDP (2002) Agriculture 2.65 Agriculture, Forestry, fishing, hunting 0.45 forestry, fishing and hunting 1.94 Agriculture, forestry, fishing and hunting 2.33 Agriculture, forestry, fishing, and hunting Mining, oil and gas 1.28 Mining, oil and gas 1.47 Mining, oil and gas 3.13 Mining, oil and gas 3.38 Utilities 1.31 Utilities 1.15 Utilities 1.32 Utilities 1.28 Construction 0.70 Construction 0.79 Construction 0.78 Construction 1.10 Manufacturing 0.65 Manufacturing 0.76 Manufacturing 0.55 Manufacturing 0.59 Wholesale trade 1.07 Wholesale trade 1.11 Retail and Wholesale trade Retail trade 1.03 Retail trade 0.94 wholesale trade Retail trade 0.76 Transportation and warehousing Accommodation and food Personal and household Professional, scientific and technical Finance and insurance Real estate and leasing Business, building and other support Information, culture and recreation Educational Health care and social assistance Public administration 0.86 Transportation and warehousing 0.91 Accommodation and food Personal and household Professional, scientific and technical Finance and insurance, real estate and renting and leasing and management of companies and enterprises Administrative and support, waste management Arts, entertainment and recreation Information and cultural industries 1.24 Educational 1.15 Health care and social assistance 1.12 Public administration 1.09 Transportation and warehousing 0.75 Accommodation and food Personal and household Professional, scientific and technical Finance, insurance, real estate and leasing Business, building and other support 0.81 Information, arts, entertainment, and 0.97 recreation 1.12 Educational 1.12 Health care and social assistance 1.13 Public Administration 1.05 Transportation and warehousing 0.92 Accommodation and food Personal and household Professional, scientific and technical Other 0.82 Information and culture Arts, entertainment, and recreation 1.13 Educational 1.00 Health care and social assistance Government 0.57 Competitive Advantage Analysis September 2005 Page 19

27 Data comparability issues limit what can be said about the large finance, insurance, real estate, and business management group. Saskatchewan seems to have a relatively low level of business, building, and support, a category that includes head offices but an average or slightly above average level in the finance and insurance group. The same comparability issues are evident in the information, culture, recreation group with Saskatchewan seeming to have a below-average level compared with others although the paid employment comparison with the midwest states and prairie provinces suggests this is only true for paid workers and not the self-employed. Generally speaking, there are relatively few differences in the two comparison measures and, in spite of the measurement difficulties, in the comparison with the two different regions. Saskatchewan s economy is dominated by agriculture and the resource sector, regardless of the measure and regardless of the comparison. It has a large, but not unduly large, public sector broadly defined to include health, education, and the crowns. The province is very weak in manufacturing and the professional, scientific, and technical and below average in construction and accommodation/food. Competitive Advantage Analysis September 2005 Page 20

28 SECTION 3 SHIFT SHARE ANALYSIS If a particular industry is growing or declining, an automatic question that arises is whether or not the decline is natural in the sense that it is growing or declining everywhere as well. Shift share analysis attempts to answer this question by breaking down the change into three components Using either GDP or employment as the measure, the shift share analysis ascribes changes in the measure over a period and within a particular region to three components. The equations below use employment as the measure and four western provinces as the comparison. For GDP, the real value of GDP is substituted for employment; for the comparison with midwestern states and provinces, the four western provinces is replaced by the values for the midwest states and prairie provinces. 1. The first component is growth or decline arising from general employment growth. If employment is generally increasing in western Canada, one would expect it to be increasing in Saskatchewan as well. This is referred to as the regional effect and is calculated for a particular sector as follows. where 92EMP SECT is the 1992 employment level in the sector in the Saskatchewan and WCEMP is the percentage change in overall employment between 1992 and 2004 in Western Canada. 2. The second component is growth or decline arising from the nature of the industry. If a particular industry group is expanding or declining generally in the comparison group, one would expect it to be expanding or declining in the Saskatchewan. This is referred to as the industry effect and is calculated for a particular sector as follows. where 92EMP SECT is the 1992 employment level in the sector in Saskatchewan, WCEMP SECT is the change in employment in the sector between 1992 and 2004 in western Canada, and WCEMP is the percentage change in overall employment between 1992 and 2004 in four western provinces. 3. The third component is the residual and is the growth or decline arising from the specifics of the industry within the region. This is referred to as the provincial effect and is the most interesting for analysis purposes. It is calculated as follows. where 92EMP SECT is the 1992 employment level in the sector in Saskatchewan, EMP SECT is the percentage change in that sector s employment between 1992 and 2004 in Saskatchewan and WCEMP SECT is the percentage change in that sector s employment between 1992 and 2004 in the four western provinces. Competitive Advantage Analysis September 2005 Page 21

29 The sum of these three effects yields overall employment growth in the sector from 1992 to The equations simply attribute the change to the three effects. Interpretation of shift share components is complex so an example may help. From 1992 to 2004, employment in the retail trade sector increased by 4,900 persons in Saskatchewan. Given overall employment growth in the four western provinces, employment in the sector would reasonably be expected to increase by 14,500 persons in Saskatchewan. If Saskatchewan had experienced the same pattern in the retail trade sector as occurred in the four western provinces, employment would have declined by 3,200 persons. That is, employment would have grown by 14,500-3,200 = 11,300. The difference between these figures is the provincial effect. That is, a decline of 6,400 persons (11,300-4,900) is necessary to explain the actual change in employment. In other words, the growth in retail trade employment was less than would be expected because of changes in the industry which were occurring throughout western Canada. The provincial effect is the unique effect for Saskatchewan. (One of those factors unique to Saskatchewan would undoubtedly be the lack of population growth which would directly affect retail sales.) Competitive Advantage Analysis September 2005 Page 22