Occupation modelling system - C4SEOCCUPATION MODELLING SYSTEM September 2014 2 Introduction This document describes the Canadian provincial occupational modelling system (POMS). POMS

OCCUPATION MODELLING SYSTEM

1 September 2014

Contents Contents ............................................................................................................................................................................................ 1

Introduction ........................................................................................................................................................................................ 2

Why LMI? ...................................................................................................................................................................................... 2

Why POMS? ................................................................................................................................................................................. 2

Data Reliability .............................................................................................................................................................................. 3

Document Content ........................................................................................................................................................................ 3

Key Occupation Labour Market Concepts ........................................................................................................................................ 4

Basic Labour Market Concepts ..................................................................................................................................................... 4

Occupation Analysis Concepts ..................................................................................................................................................... 8

Normal Unemployment ............................................................................................................................................................. 8

Labour Force Demand and Expansion Demand ...................................................................................................................... 8

Expansion Supply ................................................................................................................................................................... 10

Total Demand and Supply Changes ....................................................................................................................................... 12

Labour Market Tightness Ranks ................................................................................................................................................. 12

POMS Occupation Labour Market Indicators ............................................................................................................................. 13

Occupation Data ......................................................................................................................................................................... 15

Occupation Labour Market Analysis................................................................................................................................................ 16

Gap Analysis ............................................................................................................................................................................... 16

Source Analysis .......................................................................................................................................................................... 18

Demand Flows ........................................................................................................................................................................ 18

Supply Sources ....................................................................................................................................................................... 19

Occupation Demand and Supply Models ........................................................................................................................................ 23

Employment ................................................................................................................................................................................ 23

Labour Force Demand and Supply ............................................................................................................................................. 23

Labour Force Change ................................................................................................................................................................. 26

Appendix: NOC Occupations .......................................................................................................................................................... 29


September 2014 2

Introduction This document describes the Canadian provincial occupational modelling system (POMS). POMS is able to conduct occupation demand and supply projections for each province and roll them up to produce projections for Canada as a whole. The occupation outlooks are consistent with the C4SE provincial economic outlooks. POMS can also be accessed by C4SE clients to produce projections using their own assumptions about the determinants – drivers – of occupation demand and supply. It should be noted that the description of POMS presented below also serves as a description of the regional occupation modelling systems used by C4SE and its clients

Why LMI? The purpose of providing occupation labour market information (LMI) such as that produced by POMS is to assist organizations in ensuring that they are able to obtain sufficient numbers of employees with the right skills, at the right time, and in the right geographic location. With such information organizations can devise strategies to help them achieve their labour market goals. There are two types of analyses associated with attempting to achieve these goals. The first type focuses on examining the expected “gap” between the demand and supply of the occupations in question. Will the markets for the occupations be tight or loose in the sense of the degree of difficulty in obtaining the required employees? Will there be a shortage or a surplus of the occupations of interest? How large might these shortages or surpluses be? Does the size of the gap suggest that we need to look closely at possible sources of supply? Information regarding this type of analysis is called “Gap Analysis.” The second type of analysis looks at what is causing the changes in demand for and supply of employees. Will the demand for employees by an organization originate from the need for more employees to meet increased demand for their products or will it come from the need to replace employees that are retiring? What will be the sources of supply to meet the demand? Will the employees come from school leavers, employees from other geographic locations, other industries, or from other occupations? Information regarding this type of analysis is called “Source Analysis.”

Why POMS? The impetus for developing POMS originates from work undertaken by the C4SE to assist clients conducting occupational analyses. There are serious challenges in attempting to undertake such analyses. The most important challenge has been and continues to be obtaining reliable historical data on occupation demand and supply. The largest challenge is obtaining data on occupation supply. While there is information on occupation employment and labour force published in the Census of Canada, National Household Survey and in the Labour Force Survey (LFS) it is very difficult if not impossible to find information to enable the modelling of the components of the change in occupation supply such as retirements from the labour force and the number of people leaving school to enter the labour force. There has been some modelling and forecasting undertaken by a number of organizations across the country. Nevertheless, this modelling and forecasting has a number of shortcomings when it comes to conducting comprehensive country wide occupation analyses. Many organizations focus on specific groups of occupations in specific industries using different assumptions about the occupation drivers. Others such as provincial governments that produce occupation outlooks normally focus on just one province. These outlooks are in almost all cases based on different assumptions about the occupation drivers and at conducted at different points in time. It is very difficult for an occupation analyst to put together this diverse set of information to conduct a comprehensive and consistent analysis for the occupations of interest. The goal in the development of POMS has been to provide occupation analysts with information that will allow them to conduct occupation analyses across the provinces at the highest level of occupation detail using consistent assumptions regarding the occupation drivers. Consistent in this context means that the information is produced at the same time using the same concepts and underlying assumptions about the occupation drivers. The assumptions about the drivers as mentioned above are obtained from the latest C4SE provincial economic forecast. The occupation information provided includes that needed to conduct gap and source analyses. In POMS the approach in modelling the demand and supply side differs from that used in almost all other models. The main difference is regarding how the modelers treat workforce supply. In POMS, workforce supply and demand are interdependent – supply responds to changes in demand and changes in supply can impact demand. Almost all of the other approaches assume that workforce supply is an input to their workforce outlooks. That is, they assume that the demand and supply for the workforce are independent of each other. This approach allows them to show large persistent imbalances between supply and demand that are inconsistent with the way the world works. In the real world such large imbalances would cause wages and prices and other economic variables such interest rates, participation rates, and the exchange rate to adjust to reduce these imbalances.


3 September 2014

The occupation projections made using POMS reflect a “requirements” approach for both demand and supply. This approach starts with the C4SE macroeconomic models where workforce demand and supply adjust over time to balance aggregate labour markets. An important part of this adjustment is an “optimal” immigration approach where the federal government is the residual source of workforce supply. Under this approach there are no persistent large imbalances at the aggregate level of the economy across the country. Nevertheless, there may be temporary shortages or surpluses for some occupations over the economic cycle that will need to be taken into account in workforce planning. In the POMS outlooks there is an outlook for demand requirements and the required supply to meet these demand requirements. The demand requirements are changes in employment, people retiring from the workforce, and people dying. The sources of supply to meet these requirements are young people entering the workforce after finishing school, both international and interprovincial net in-migrants, and other sources such as people changing occupations and deciding to enter the labour force because of more job opportunities and higher wages. While there are no major persistent labour market imbalances in the POMS outlooks, POMS does employ a labour market tightness ranking approach that takes into account the fact that the supply requirements computed may not be achieved. It attempts to identify occupations that may be difficult for organizations to find in the future. Occupations with relatively strong demand growth, for example, may be more difficult to find than those where demand growth is weaker. Moreover, occupations where supply requirements are largely met through migration may be at risk if the federal government does not accommodate these requirements through additional immigration, or, Canadian workers do not wish or are not available to move to the particular location in question. There may also be some tightness issues over economic cycles and for specific occupations. This tightness is incorporated in the ranking approach.

Data Reliability It should be noted that the reliability of the information produced by POMS declines with the size of the occupation labour force in the provinces. The projections for occupations with few members – say less than 100 – should be treated with caution. There are not very many of the 500 occupations with less than 100 in the larger provinces, but a number of them are found in the smaller provinces. While the numbers produced for employment and labour force for these occupations may be useful in suggesting trends, there is certainly a danger in conducting analysis of precise supply-demand gaps for them.

Document Content The next section of this document describes the labour market concepts used in POMS. This description includes that for labour

market indicators published by Statistics Canada as well as that used in POMS and for occupation analysis. The third section

contains an example of how the information produced by POMS can be used to conduct an occupation analysis. The final

section is a more technical one that describes the approach used in POMS to provide this information. The Appendix contains a

list of the industries and 2011 National Occupation Categories (NOCs) covered in POMS.


September 2014 4

Key Occupation Labour Market Concepts Before describing the approach used to produce information for occupation labour markets it is important to present the key concepts that are part of this approach. Without an understanding of these concepts it will be difficult to use and interpret the information produced by POMS. The types of concepts employed are examples of stock and flow variables. In examining possible shortages (gaps) of occupations the concern is with stock (level) variables such as the labour force. To identify sources of changes in the demand and supply of occupations the approach focuses on flow (change in stocks) variables such as retirements and migration. Stocks refer to variables measured at a point in time. An example of a stock variable is the size of the population on July 1, 2012. Flows refer to changes in variables measured between two points in time. The number of persons moving into Canada on a permanent basis from July 1, 2011 to June 30, 2012, which is called immigration, is an example of a flow variable. It is one of the flow variables along with births and deaths that measure the change in the size (stock) of the population between two points in time.

Basic Labour Market Concepts The labour market information in the modelling system is derived from that provided by Statistics Canada in their Labour Force Survey and Census of Canada. Each month Statistics Canada publishes information about the labour force, employment, unemployment, the unemployment rate, source population, and the labour force participation rate for various jurisdictions across the country. The definitions of these labour market variables are as follows:

Employment (stock): number of people who are working at a particular point in time. This number includes both full and part time employees;

Labour force (stock): number of people working (employment) plus the number of people actively looking for work at a particular point in time;

Unemployment (stock): persons who are actively looking for work but are unable to find it at a particular point in time – calculated as labour force minus employment;

Unemployment Rate: percentage of the labour force that is unemployed at a particular point in time – calculated as unemployment divided by labour force multiplied by 100;

Source Population (stock): – number of persons in the population aged 15 years and over that is able to work. The source population excludes persons in institutions such as prisons and hospitals or those that are ill or disabled and unable to work; and

Labour Force Participation Rate: the percentage of the source population that is in the labour force – labour force divided by source population multiplied by 100.

It is important to emphasize the part of the definition of the labour force that refers to “actively looking for work.” People who are not employed and who are not actively looking for work are not considered to be part of the labour force. Such persons are not available to organizations looking for workers and, therefore, are not counted as part of labour supply. The labour force is the measure of labour supply that is used in the occupational modelling system. For the economy as a whole it is defined algebraically as the product of the participation rate and the source population:

Labour Force equals Labour Force Participation Rate divided by 100

multiplied by the Source Population As a result, the supply of labour as measured by the labour force changes in response to changes in the labour force participation rate and the source population. Changes in the source population are caused by the same factors as those that cause changes in the overall population. The change in the population in a province, for example, is equal to births minus deaths plus net in-migration to the province. In a particular year births do not influence the source population since the latter variable is comprised of population 15 years of age and over. Nevertheless, births will impact the source population 15 years from the particular year in question. There is a number of determinants of the labour force participation rate for the economy as a whole. The most important one is the need for individuals to obtain income to finance their purchases of goods and services. This need differs across the age groups in the population.


5 September 2014

Young people participate relatively less in the labour force because they attend school to obtain the knowledge and skills required when they become more permanently (strongly) attached to the workforce. To finance their purchases they may borrow funds from financial institutions or their parents or find part-time work. Older people also participate relatively less – starting after about 50 years of age – as they decide to retire from the labour force. They finance their purchases through pensions or savings made through working at earlier ages. Some of them continue to work, but relatively more on a part-time basis. Changes in labour compensation and sources of income also influence the decision to enter or leave the labour force. Higher after-tax wages will tend to have the impact of increasing the labour force participation rate. Increased non-labour income such as from investments or government transfer payments will reduce the participation rate as there is less need to work to earn income. The labour force participation rate is also impacted by social and cultural factors. The decision by women to increase their participation in the labour force, for example, has been and continues to be a very important determinant of increases in the labour force. Examples of the evolution of employment, labour force and the unemployment rate for Canada as a whole are shown in Figure 1. Employment and labour force are measured on the left axis in thousands of persons and the unemployment rate is measured on the right axis in percentage terms. The unemployment rate represents the percentage gap between the labour force (top line) and employment (the line shown under the labour force). As can be seen from Figure 1, the unemployment rate, apart from exhibiting a downward trend, has tended to cycle up and down over the period. This cycling reflects changes in the level and growth of economic activity. The unemployment rate, for example, rose from 1989 to 1993 during a recession; it fell in the recovery from the recession; it rose again around 2000 with another economic slowdown; it declined in the following recovery; and then increased in the 2009 recession.

Figure 1

Employment (000s), Labour Force (000s) and Unemployment Rate (%), Canada

1981-2010

Source: Statistics Canada, Labour Force Survey The rise and fall of the unemployment rate represents a tightening and loosening of the labour market as demand measured by employment increases relative to supply as measured by the labour force. The downward trend in the unemployment rate suggests an upward trend in the tightness of the labour market. The latter trend has been suggested to be partly a result of the aging of the workforce.

0

2

4

6

8

10

12

14

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

1981 1984 1987 1990 1993 1996 1999 2002 2005 2008

Employment (LHS) Labour Force (LHS) Unemployment Rate (RHS)


September 2014 6

Figure 2 shows the labour force participation rate – left axis – and the unemployment rate – right axis. The labour force participation rate, like the unemployment rate, has cycled over the period 1981 to 2010 and for related reasons. The cycles nevertheless are inversely related. This relation is a result of what is called the “Discouraged (Encouraged) Worker Effect.” The latter effect states that as the economy goes into a recession or a noticeable growth slowdown and the unemployment rate rises, the percentage of the source population looking for work falls, as people find it difficult to find work and drop out of the labour force – quit looking for work. Conversely, when the economy is picking up and entering a boom period the percentage of the source population looking for work rises as it is easier to find work and wages are increasing. The slight upward trend in the participation rate over the period would appear to reflect in part the tightening of the labour market.

Figure 2 Labour Force Participation Rate and Unemployment Rate (%), Canada

1981-2010

Source: Statistics Canada, Labour Force Survey

As mentioned above, an aspect of the labour force participation rate that is important for occupation labour market analysis is its age and sex distribution. Figure 3 shows estimates of the participation rates by age and sex for Canada for 2010 for ages 15 to 70 and over. As can be seen from the figure, the rates for males are higher than those for females and the rates for both males and females differ across age groups. For ages 15 to 30 the rates rise with age as more and more people enter the labour force on a more permanent basis after leaving school. From 30 years of age to the late 40s the rates are relatively constant. After 50 the rates decline at an increasing rate with age as workers retire from the labour force.

Figure 4 shows the aggregate – all ages – participation rates for males and females over the 1981 to 2010 period. As can be seen from the figure, the participation rate for women has risen substantially since 1981 more than offsetting the decline in that observed for males.

0

2

4

6

8

10

12

14

62

63

64

65

66

67

68

1981 1984 1987 1990 1993 1996 1999 2002 2005 2008

Participation Rate (LHS) Unemployment Rate (RHS)


7 September 2014

Figure 3

Labour Force Participation Rates by Age and Sex, Canada, 2010

Source: Statistics Canada and C4SE

Figure 4 Labour Force Participation Rate, Males and Females, Canada

1981-2010

Source: Statistics Canada, Labour Force Survey

0

10

20

30

40

50

60

70

80

90

1981 1984 1987 1990 1993 1996 1999 2002 2005 2008

Females Males

0

10

20

30

40

50

60

70

80

90

100

15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69

Females Males


September 2014 8

Occupation Analysis Concepts The concepts described above are those that are published by Statistics Canada and used in the media and by the public in regular conversations about the state of the labour market. In analyzing occupation labour markets additional concepts are required. These concepts define what is meant by the demand for and supply of occupations as well as the sources of changes in occupation demand and supply. The occupation modelling system on the demand side focuses on the workforce required by private and public organizations in the economy. Workforce here refers to the number of persons required, not the number of hours required from these persons. On the supply side, the available workforce, as measured by the labour force, is what is of interest for the modelling system. While it would be desirable to use hours to measure supply and demand such data are not available. The key concepts for an occupation are as follows:

• Labour Force Demand (stock): employment plus the normal level of unemployment for an occupation; • Normal Unemployment (stock): unemployment normally observed for an occupation because of the nature of the work

and the industries in which it is primarily employed; • Labour Force Supply (stock): the labour force as described above; • Excess Labour Force Supply (stock): labour force supply minus labour force demand; • Normal Unemployment rate: the percentage of supply that is normally unemployed; • Expansion Demand (flow): the change in occupation demand (stock) as defined above; • Expansion Supply (flow): the change in occupation labour force (stock) as defined above; • Total Demand Change, Job Openings, or Job Requirements (flow) – is the sum of expansion demand (flow),

retirements (flow), and deaths (flow); • Total Supply Change (flow) – is the sum of new entrants (flow) and other sources of supply (flow); and • Labour Market Tightness Rank – is a number that reflects the degree of tightness (difficulty of finding workers) in the

labour market. The first four concepts are stocks and what are of interest when asking whether or not there is sufficient supply to meet demand for an occupation. The next four variables are flows and are of interest in describing the sources of change in the demand for and supply of an occupation.

Normal Unemployment While it would be desirable for there to be no unemployment in the economy such a situation is not possible. Given the nature of the labour market it is necessary that there be some unemployment to facilitate its proper functioning. The latter unemployment is what is defined as normal unemployment. There are a number of different types of unemployment:

• Seasonal unemployment: many activities in the economy such as construction have a seasonal component where fewer or more workers are required at different times during the year. For these occupations a higher level of the labour force in relation to employment is required to meet peak demand for workers;

• Frictional unemployment: there is always a number of people between jobs either as they search to improve their careers or move to a different geographic location;

• Structural unemployment: as the economy changes over time there will always be some mismatch between the skills required and those possessed by workers in the local economy; and

• Cyclical unemployment: is the unemployment that is associated with recessions and recoveries as the economy goes through economic cycles.

Normal unemployment refers to the first three types of unemployment. Normal unemployment rates differ across occupations reflecting differences in the seasonal and other aspects of a job. Managers, for example, have relatively low rates while those for construction related trades are relatively high as the work is seasonal and more labour force is required to meet peak levels of economic activity.

Labour Force Demand and Expansion Demand Figure 5 shows an example of labour force demand for an occupation. It is comprised of employment and normal unemployment. The latter component represents about 4 percent of labour force demand for this occupation.


9 September 2014

Figure 5

Labour Force Demand

Source: C4SE Expansion demand refers to changes in labour force (labour force demand) requirements for the occupations. The sources of change in expansion demand are the change in employment and the change in normal unemployment. The change in normal unemployment is directly related to changes in employment through the normal unemployment rate. If the normal unemployment rate is 5 percent, for every 100 new employees required for an occupation, an additional 5 would be needed for the labour force to keep the unemployment rate at 5 percent. The 100 new jobs add 100 new persons to employment and 105 persons to the labour force. Figure 6 shows the components of expansion demand for the occupation used in Figure 5. As can be seen from the figure, expansion demand can be both positive and negative as it represents the change in labour force demand shown above. The change in normal unemployment is always in the same direction as employment as described above.

Figure 6 Expansion Demand

Source: C4SE

-200

-150

-100

-50

0

50

100

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020

Employment Change Normal Unemployment Change

0

500

1000

1500

2000

2500

3000

3500

2008 2010 2012 2014 2016 2018 2020

Employment Normal Unemployment


September 2014 10

Expansion Supply Expansion supply is the change in an occupation’s labour force. This change is comprised of the following parts: Additions to Supply:

New Entrants In-Migration Other In-Mobility

Reductions in Supply:

Deaths Retirements Out-Migration Other Out-Mobility

In modelling the occupations, migration and mobility are expressed in net terms: net in-migration equals in-migration minus out-migration and net other in-mobility equals other in-mobility minus other out-mobility. Availability of data is an important reason for using this net approach.

New Entrants The number of new entrants to the labour force refers to persons entering the labour force from the population in the 15 to 30 age group as described above. They represent additions to the labour force. Migrants or occupations such as managers or supervisors that require related labour market experience are not included in this category. This concept is meant to refer to persons that enter the labour force either for the first time after completing their education, or, if previously working part time while receiving their education, as they complete their education and start to work in their chosen occupation. While related to the concept of school leavers that is often used in occupation analysis, it is not the same concept. It also does not include people completing their apprenticeship, as these people normally apprentice after entering the labour force. An example of how new entrants differ from school leavers is a student who joins the labour force for 3 months in the summer while obtaining his or her education but does not work during the school year. This student would be counted as 0.25 persons – 3 divided by 12 months – in the labour force for the year as a whole. When they begin to work 12 months a year after completing their education they are counted as 1 person in the labour force – their annual participation rate in the labour force jumps from 25 percent to 100 percent. New entrants would capture this latter labour force increase of .75, not 1 implied by school leavers. The increase in the participate rate from ages 15 to 30 shown in Figure 3 above reflects this type of activity.

Retirements Retirements subtract from the labour force for an occupation. In occupation analysis, the concept of retirement is meant to refer to those persons who leave the economy’s labour force – they no longer work. Often the concept of retirement is not interpreted in this manner by the public. Some people retire from their job and then take up work in another job either in the same occupation or another occupation. In this case the person has not left the economy’s labour force. If they work in another occupation then they represent inter-occupation mobility – movement between occupations such as a person in the construction trades who retires from construction and works in a restaurant as a waiter on a part time basis. One of the problems encountered in occupation analysis is that information on retirements for occupations is normally not provided on the basis required. For example, it has recently been stated in a story in the paper that the average age at retirement for teachers in Ontario is 59. Does this mean that teachers no longer work in the labour force or that they are “officially” retired from teaching – some teachers qualifying for retirement may retire and then take supply teacher positions or a job in a different occupation? To the extent that the reported retirement ages refer to the latter situation, retirements from the overall labour force will be overestimated. This does not affect the size of the overall labour force because in POMS the sum of the individual occupation labour forces is required to add to that for the economy as a whole – the latter is an input to POMS. As a result, the overestimation – or underestimation if the retirement rates are too high – of retirements is allocated to net in-mobility – largely inter-occupation mobility – which is described below. The economy’s total labour force is computed using labour force participation rates by age and sex. These rates implicitly show the percentage of people retiring from the labour force. As mentioned above in the discussion on participation rates, participation rates by age fall starting in the late 40s reflecting retirements from the labour force. As more and more people move into older age groups the number of retirements increase.


11 September 2014

Deaths Deaths refer to those occurring from all causes, not just on-the-job deaths. Deaths subtract from the labour force.

Net In-Mobility Net in-mobility is defined as the sum of net in-migration and other net in-mobility. This component of labour force change measures the change in the existing labour force that is required because withdrawals from the labour force – deaths and retirements – are not equal to new entrants to the occupation’s labour force.

Net In-Migration Net in-migration refers to persons moving into or out of a geographic area to take or find a job. Positive net in-migration adds to the labour force while negative net in-migration subtracts from the labour force.

Other Net In-Mobility Other net in-mobility refers to net additions to an occupation’s labour force from such sources as persons changing occupations – inter-occupation mobility – and changes in labour force participation rates for social or cyclical reasons as described above. Figure 6 illustrates the components of expansion supply for the example occupation. The bars represent the components while the line represents the value of expansion supply – change in labour force. The new entrant component is always positive – on or above the zero line. The retirements and death component is always negative – on or below the zero line. Net in-mobility can be either positive or negative. As mentioned above net in-mobility is the amount required to offset the difference between new entrants and retirements and deaths to explain expansion supply. In Figure 6, new entrants trend downward slightly over the period. Retirements and deaths trend upwards. The latter trend reflects the aging of the population. The former trend reflects relatively low birth rates and, as a result, fewer young people entering the labour force. The large change in the labour force in 2008 and generally after 2011 requires positive net in-mobility to the labour force as the number of new entrants less deaths and retirements is insufficient to account for the labour force supply change. Both migration and other net in-mobility contribute to the amount of net in-mobility.

Figure 6 Components of Expansion Supply (Labour Force Change)

Source: C4SE

-150

-100

-50

0

50

100

150

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020

Deaths & Retirements New Entrants

Net In-Mobility Labour Force Change


September 2014 12

The negative values of expansion supply in 2009 and 2010 – decline in the labour force supply – are a result of reductions in labour force demand. A major component of this decline is net in-mobility that results from a reduction in participation rates as described for the Discouraged Worker Hypothesis above. The positive value of net in-mobility from 2011 to 2013 is also partly a result of this effect in reverse. Increased in-migration also makes a positive contribution to net in-mobility during this period.

Total Demand and Supply Changes Over the long run it is generally expected that labour force demand will be equated to labour force supply. In this case the change in demand will also be equal to the change in supply:

Expansion Demand equals Expansion Supply

Substituting the components of expansion supply:

Expansion Demand equals New Entrant minus Retirements and Deaths plus Net In-Mobility

Rearranging the equation yields:

Expansion Demand plus Retirements and Deaths equals New Entrants plus Net In-Mobility The components on the left hand side of this equation represent total demand change while those on the left hand side represent total supply change in POMS. Expansion demand represents net “new” jobs, while retirements and deaths represent “replacement” demand. It should be noted that all retirements and deaths for an occupation need not be replaced if expansion demand is negative. For example, suppose expansion demand is -1000 and retirements and deaths are 2000. The total demand change is 1000. In this case only 1000 persons of the 2000 who died or retired need to be replaced. New entrants always add to the labour supply while net in-mobility can add or subtract from labour supply. The components of net in-mobility are really found on both sides of the equation as mentioned above. If net in-mobility is negative it is part of replacement demand, when it is positive it is part of supply. If people leave an occupation they may need to be replaced when the total demand change remains the same. Net in-mobility is usually negative in response to similar changes in expansion demand and vice versa.

Labour Market Tightness Ranks The tightness ranks included with the POMS information are qualitative measures of the possible tightness of the labour market. There are three ranks measured from 1 to 3. A 1 represents a loose market characterized by excess supply, a 2 represents a normal market – a type of market situation that does not show extremes regarding the difficulty or ease of finding workers – and a 3 represents a tight labour market characterized by excess demand. Four ranking measures are adopted with POMS. The first measure, the gap rank, is created using actual and normal unemployment rates for an occupation. If the actual unemployment rate is in excess of 20 percent above the normal rate the tightness ranks is 3 (tight). If it is less than 20 percent below the normal unemployment rate the rank is 1 (loose). Otherwise the rank is 2 (normal). Analysts can compute their own measure of tightness by specifying the bands around the normal unemployment rate. The requirements approach adopted in POMS would not be expected to show large or widespread labour market imbalances as supply responds to demand. The gap rank on average over time will always be 2 – showing a relatively balanced market. From the point of view of assessing the risk for organizations attempting to meet their labour requirements the gap rank provides little information in the context of POMS unless the gap for an occupation is relatively large. To better assess the risk two additional rank measures are employed. These measures assess the “relative” labour market tightness of the occupations.


13 September 2014

The second ranking measure – demand rank – assesses the state of the relative demand for occupations. This measure is calculated using the growth in job openings or requirements for an occupation as a percentage of its labour force as shown in Table 1 below. The higher is this value the more difficult it will be to obtain these occupations given the available supply. The rank for this measure is computed as follows:

Estimate the mean and standard deviation of the growth value over all occupations;

For each occupation find where its job-openings as a percentage of labour force value falls in the overall distribution of the values;

If the value for an occupation is within 1 standard deviation of the mean the rank is set at 2;

If the value for an occupation is greater than 1 standard deviation above the mean its rank is 3; and

If the value for an occupation is greater than 1 standard deviation below the mean its rank is 1; The third ranking measure – supply rank – focuses on the migration component of supply as a source of risk for finding the required workforce. This measure is calculated using the percentage of an occupation’s labour force that needs to be sourced through migration. The higher is this percentage the more difficult it will be to find the workforce. The rank for this measure is calculated in the same manner as that for the second measure. A fourth ranking measure is computed that is a weighted average of the previous three measures. The default weights for the three ranks are 40 percent for each of the demand and supply ranks and 20 percent for the gap rank. Users of the outlook can set their own weights for these measures to compute the fourth ranking measures.

POMS Occupation Labour Market Indicators Figure 7 illustrates the type of occupation indicators produced by POMS. Many of these indicators have been described above. Along with some of the indicators presented is the percentage change or level change in them from year to year. For the labour force change components a negative sign is placed in front of deaths and retirements as they represent reductions from the labour force. The death rate is presented as the number of deaths per 1000 persons in the labour force. For the other flow variables such as retirements and new entrants the percentage of the labour force (from the previous year) represented by the indicators is shown. In the case of retirements this percentage is the labour force retirement rate for the occupation. The percentage of the labour force provides some idea of the importance of the indicators to labour force change. For example, the percentage for retirements is 2.9 percent suggesting that that almost 3.0 percent of the labour force may need to be replaced that year. The average age of the labour force is also provided. The higher the average age the larger is likely to be the number of deaths and retirements as a percentage of the labour force for an occupation. Under gap analysis excess supply, which is the difference between labour force supply and labour force demand, is presented. The negative value suggests that supply is not meeting demand as defined in POMS and the occupation’s labour market is or will be tight. The unemployment rate gap is the difference between the actual and normal unemployment rates. While the unemployment rate is positive, the unemployment rate gap is negative as the unemployment rate is below the normal unemployment rate, which suggests that it may be difficult to find employees in this occupation. Nevertheless, the tightness ranks in the table are all 2, suggesting that the labour market is balanced. Under flow analysis the components of total demand change of 794 are comprised of 284 new employees and normal unemployment – expansion demand – and 510 deaths and retirements that need to be replaced. The change in supply of 810 is met through 347 new entrants and 463 from net in-mobility of employees. As can be seen from the numbers in Figure 7, the latter is met by 112 migrants and 350 other net in-mobility – there is some rounding error in the presentation of these numbers.


September 2014 14

Figure 7

Example Occupation Table

Labour Force Demand 21562

% Change 0.5

Change 110

Employment 20112

% Change 0.5

Change 103

Normal Unemployment 1450

Labour Force Supply 21542

% Change 1.2

Change 262

Deaths 68

Death Rate (Per 1000 Persons)

Retirements 304

% of Labour Force 1.4

New Entrants 462


Net In-Mobility 172


Net In-Migration 247


Other Net In-Mobility -75

% of Labour Force -0.3

Average Labour Force Age 41.03

Gap Analysis

Excess Supply -21

Unemployment Rate 6.64

Normal Unemployment Rate 6.73

Unemployment Rate GAP -0.09

Weighted Labour Market Tightness Rank (1-3) 2

Demand Rank 2

Supply (Migration) Rank 2

Unemployment Rate Gap Rank 2

Flow Analysis

Total Demand Change Sources 482


Expansion Demand 110

Deaths & Retirements 372

Total Supply Change Sources 634

New Entrants 462

Net In-Mobility 172


15 September 2014

Occupation Data The HRSDC NOC system of occupation definitions is adopted for the modelling system – http://www5.hrsdc.gc.ca/noc/english/noc/2011/welcome.aspx. There are 4 levels of NOC occupations. An example of these levels is:

• 0 Management occupations – 1 digit o 00 Senior management occupations – 2 digit

001 Legislators and senior management – 3 digit

0011 Legislators – 4 digit The occupational modelling system works at the 4-digit level of occupation aggregation where there are 500 occupations. The occupation concepts described above are measured on a Place of Residence basis. The concepts refer to persons living in a province. Persons living in a province may have a Place of Work in the province or outside the province. The occupation modelling system focuses on the demand for workers in the province. To the extent that occupations work largely outside the province or are dominated by workers living outside the province, it will present a less accurate picture of Place of Residence measures for the province. The data for the modelling system are sourced from the Census of Canada or lately the National Household Survey (NHS) and the Labour Force Survey (LFS). They are measured on an LFS basis for the occupations as a whole. That is, the total labour force and total employment are equal to the LFS values each year. Nevertheless, the occupation data do not match the LFS occupation data published by Statistics Canada. The latter data have too small a sample size to provide reliable estimates of the occupations, particularly on a provincial basis. To estimate the occupation employment data, an occupation’s share of Census (NHS) employment of Census (NHS) total employment in an industry is applied to the corresponding LFS industry employment. The occupation labour force data are estimated in a two-step procedure. In the first step the ratio of Census (NHS) labour force to Census (NHS) employment is applied to the LFS equivalent employment created above. Next, the resulting labour force estimates are normalized to total LFS labour force. For some occupations sample sizes were such that either information were not available or did not make sense because of Statistics Canada’s random rounding procedure for data. In the latter case employment exceeded labour force, which is not possible. The solution to this situation was to set the labour force 3 percent above employment – the unemployment rate is three percent. It should be noted that Statistics Canada warns users of their data about the poor reliability of data for occupations with few observations, which is particularly the case for provinces with relatively small populations. The employment share coefficients for the occupation employment part of the model were estimated as the ratio of an occupation’s employment in an industry to the industry’s total employment for the latest Census year. The normal unemployment rates are estimated using historical data on actual unemployment rates for the occupations and judgement regarding what the value of the rate should be to ensure properly functioning labour markets. The ratio of historical averages of the actual rates by occupation to that for the economy as a whole are first calculated. These ratios are then applied to the estimated economy wide rate over the projection period. In some cases minimum values for the rates were applied when it was felt that the rates were too low – less than 3 percent. The unemployment rate data are obtained from Statistics Canada’s LFS. The retirement rates are computed from data on the median age of retirement for the occupations. The latter data are obtained from information previously published by Statistics Canada. The estimation procedure involves fitting a retirement distribution around the median age at retirement using retirement rates by age. Because the Statistics Canada information is provided only at the 2-digit NOC level, other sources such as industry information are employed where possible. Data on retirement rates at the 4-digit level from Human Resources, Skills, and Development Canada are also used to calibrate the retirement rates. The death rates are derived from information published by Statistics Canada. It is assumed that the rates for each occupation are the same as those for the general population. Historical data are not available for the components of labour force change. These components are estimated in the past and into the future using the models’ equations for them.

http://www5.hrsdc.gc.ca/noc/english/noc/2011/welcome.aspx


September 2014 16

Occupation Labour Market Analysis The POMS produces information that can be used to help Labour Market Information (LMI) analysts to identify significant supply-demand gaps across the occupations and the possible sources of supply to remove these gaps. An example of how the LMI produced by POMS can be used is provided below. This example employs gap and source w analysis both within a province and across the country. In the example a Human Resource Analyst wishes to assess the existing and future state of the labour market for heavy equipment operators as the analyst’s firm is planning to expand its oil sands operations in Alberta. The company would prefer to hire workers with related experience in the oil and gas industry in Alberta, but accepts that it may need to look to other industries and other provinces to obtain the operators. The oil sands industry is a relatively new one and there is unlikely to be significant local supply available given the planned expansion of the oil sands industry in the province over the next 10 years. Heavy equipment operators are found under NOC 7521. According to the NOC: “Heavy equipment operators operate heavy equipment used in the construction and maintenance of roads, bridges, airports, gas and oil pipelines, tunnels, buildings and other structures; in surface mining and quarrying activities; and in material handling work. They are employed by construction companies, heavy equipment contractors, public works departments and pipeline, logging, cargo-handling and other companies.”

The analysis illustrated below examines a broader category than the exact type of person required for the oil sands firm, but does provide an indication of the labour market situation for persons that possess similar qualifications. This issue is similar for other NOC occupations where the occupations are often broadly defined. In the case of carpenters, for example, organizations may be looking for framers that work in residential construction while the carpenter NOC includes framers and carpenters that work building scaffolding largely in non-residential construction.

Gap Analysis Gap analysis is used to assess the state of tightness in the labour market for operators. If there is generally an excess supply of operators, it should be relatively easy to find them. Otherwise, more effort will be required in the search for them. The measures used to conduct gap analysis include quantitative ones that employ a measure of excess supply including the use of actual and normal unemployment rates and a qualitative one that uses a ranking system regarding the perceived degree of excess demand or supply. Table 2 shows the information used for gap analysis for heavy equipment operators in the province. As can be seen from the table, a gap opens up between demand and supply in the medium term with supply falling below demand, this gap reaches 202 by 2019 – excess supply is negative at -202. To get an idea of how serious a problem this situation is – the degree of market tightness in the province – Figure 12 shows the actual and normal unemployment rates for operators over the period.

Table 2

Gap Analysis Information for Heavy Equipment Operators, Alberta

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Labour Force Demand 21562 22584 23069 23606 24171 24718 25251 25710 25804 25792 25816

Labour Force Supply 21542 22486 23073 23575 24081 24563 25050 25517 25672 25704 25770

Labour Force Excess Supply -21 -97 4 -30 -90 -155 -202 -192 -133 -88 -45

Unemployment 1430 1422 1555 1557 1536 1508 1497 1537 1603 1647 1691

Unemployment Rate 6.6 6.3 6.7 6.6 6.4 6.1 6.0 6.0 6.2 6.4 6.6

Normal Unemployment Rate 6.7 6.7 6.7 6.7 6.7 6.7 6.7 6.7 6.7 6.7 6.7

Labour Market Tightness Rank (1-3) 2 2 2 2 2 2 2 2 2 2 2


17 September 2014

Figure 12 Actual and Normal Unemployment Rates for Heavy Equipment Operators, Alberta

Unemployment rates near the normal unemployment rate – the value of the unemployment rate that is on average observed for operators in Alberta – suggests a normal labour market, while unemployment rates noticeably above or below the normal rate suggests either a loose or tight market, respectively. Figure 12 contains a band that defines the values of the unemployment rate that represent a normal labour market as defined by the POMS ranking system. The chart suggests a normal market situation. The actual unemployment rate drops below the normal rate over the medium term of the forecast, but not by a significant amount. The labour market tightness rank of 2 over the forecast along with the unemployment rate gap suggests a normal rate of difficulty in finding operators in Alberta during this period. Given that it appears that the labour market for operators in Alberta will be normal it may not be necessary to go outside the province to find them. If they were to look outside the province however, the degree of difficulty in this effort will depend on the demand-supply situation in other provinces. To assess this situation it is necessary to create versions of Figure 12 for each province. Rather than displaying the additional figures here, Table 3 shows the labour market tightness rankings for operators across the provinces. In addition to using the rank values, the table employs different colours to represent the values. Red represents excess demand, green a normal situation for the labour market, and blue (not seen in this occupation) a situation of excess supply. This approach allows decision makers and other observers to easily understand and comment on the analysis. As can be seen from Table 3, most provinces will experience normal labour market tightness for heavy equipment operators in the future. The excess demand rankings in British Columbia, Manitoba, New Brunswick, Prince Edward Island and Newfoundland & Labrador in the short to medium term likely reflect upcoming major projects in these provinces, placing pressure on the demand for operators.

Table 3

Labour Market Tightness Rankings for Heavy Equipment Operators, All Provinces

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Unemployment Rate Normal Unemployment Rate

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

British Columbia 2 2 3 3 3 2 2 2 2 2 2

Alberta 2 2 2 2 2 2 2 2 2 2 2

Saskatchewan 2 2 2 2 2 2 2 2 2 2 2

Manitoba 2 2 2 3 3 2 2 2 2 2 2

Ontario 2 2 2 2 2 2 2 2 2 2 2

Quebec 2 2 2 2 2 2 2 2 2 2 2

New Brunswick 2 2 2 3 3 3 2 2 2 2 2

Nova Scotia 2 2 2 2 2 2 2 2 2 2 2

Prince Edward Island 2 2 2 2 2 2 3 2 2 2 2

Newfoundland & Labrador 2 3 2 2 2 2 2 2 2 2 2


September 2014 18

Source Analysis The purpose of flow analysis is to identify the possible sources of supply and demand changes for the occupations in question. How much will demand and supply grow (decline) and what will be the sources of this growth (decline)? This analysis is illustrated below for the operators. Table 4 shows the sources of demand and supply change for heavy equipment operators in Alberta over the 2013 to 2023 period.

Demand Flows The sources of total demand change in POMS refer to expansion demand and retirements and deaths. Expansion demand measures the change in employment and the associated normal unemployment. It is often referred to as “new jobs” but can also refer to “lost jobs” as firms reduce their need for employees. Retirements and deaths are a measure of replacement demand.

Table 4 Demand and Supply Changes for Heavy Equipment Operators, Alberta

Figure 13 shows the two components of demand change for heavy equipment operators in Alberta. It can be seen that while expansion demand makes the largest contribution to demand change in the short term, retirements and deaths are an increasingly important source, contributing nearly the entire total demand change after 2020.

Figure 13

Sources of Total Demand Change for Heavy Equipment Operators, Alberta

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Total Demand Change 482 1405 886 949 986 976 970 901 546 443 484

Expansion Demand 110 1021 485 537 565 547 534 458 95 -12 24

Deaths & Retirements 372 384 401 412 421 429 437 443 451 456 461

Total Supply Change 634 1328 987 915 927 911 924 911 605 488 527

New Entrants 462 480 479 477 477 485 495 504 512 524 537

Net In-Mobility 172 849 508 438 449 426 428 407 93 -36 -10

Net In-Migration 247 692 450 489 487 471 490 431 115 -51 -15

Net Other Mobility -75 157 58 -51 -38 -45 -62 -25 -22 15 4

-200

0

200

400

600

800

1000

1200

1400

1600

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Expansion Demand Replacement Demand (Deaths & Retirements)


19 September 2014

Figure 14 shows the sources of total demand change in the rest of Canada. As can be seen from this figure, replacement demand, as represented by retirements and deaths, accounts for almost all the change in demand for heavy equipment operators over the period. Expansion demand for operators is negative after 2020 across the rest of the country.

Figure 14

Sources of Total Demand Change for Heavy Equipment Operators, Rest of Canada

It is useful to examine the industry composition of expansion demand over the period to see what is accounting for the relatively strong expansion demand in Alberta. Table 5 shows the employment levels for operators in 2013 and 2023 along with the change over the period with a focus on heavy equipment operators working in the industrial sector for Alberta and the rest of Canada. More detail is available for the industries found under the private and public services, but is not presented here. It is the industrial sector that employs the largest number of heavy equipment operators in Alberta, the largest portion of which works in the construction industry. In the rest of Canada the construction industry also dominates, with the largest proportion of operators also working in the other mining industry. Over the next 10 years the largest change in employment for this occupation observed in Alberta is dominated by the construction industry. This industry accounts for almost all of the increase in heavy equipment operator employment in the province. The rest of the industrial sector will show much smaller increases with some industries showing a decline in the number of workers. The construction industry also accounts for the largest part of the increase in employment for operators in the rest of Canada, increasing by about 1400 over the 2013 to 2023 period. Private and public services employment increases almost 1400 over the same period.

Supply Sources The sources of supply change shown in Table 4 are new entrants and net in-mobility, the latter consisting of net in-migration and net other mobility. Net other mobility includes workers moving from other occupations – inter-occupation mobility – and other factors that impact the participation rate of the population in the labour force such as the availability of higher wages and decisions to postpone retirement. Figure 15 shows the supply change components for Alberta and Figure 16 the corresponding components for the rest of Canada. As can be seen from these figures the total supply changes generally follow those for demand shown in Figures 13 and 14. New entrants are a less volatile source of supply than the other two components. In the case of Alberta, strong demand for operators draws young people into the occupation over the period. For the rest of Canada the number of new entrants remains largely unchanged in line with the weak employment growth shown in Table 5.

-2000

-1000

0

1000

2000

3000

4000

5000

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Expansion Demand Replacement Demand (Deaths & Retirements)


September 2014 20

Table 5

Employment of Heavy Equipment Operators, Alberta and Rest of Canada

As the previously tight labour market in Alberta begins to ease, less net in-migration is required to meet the province’s supply needs for operators.

Figure 15

Sources of Total Supply Change for Heavy Equipment Operators, Alberta

Change Change

2013 2023 2013-2023 2013 2023 2013-2023

Total 12015 13414 1399 77818 83404 5586

Industrial 10332 11476 1144 65180 69397 4217

Oil and gas extraction 358 365 7 4101 4509 408

Support activities for mining and oil and gas extraction 247 239 -8 419 409 -10

Mining 1503 1523 20 5270 6750 1480

Utilities 27 31 4 550 633 83

Construction 6650 7830 1180 43852 45963 2111

Wood product manufacturing 321 297 -24 1662 1666 4

Paper manufacturing 148 135 -13 638 613 -25

Petroleum and coal products manufacturing 0 0 0 446 369 -77

Chemical manufacturing 23 21 -2 313 324 11

Non-metallic mineral product manufacturing 113 143 30 806 804 -2

Primary metal manufacturing 56 53 -3 1110 1037 -73

Fabricated metal product manufacturing 17 18 1 338 362 24

Machinery manufacturing 0 0 0 387 456 69

Other Industrial 869 821 -48 5288 5502 214

Private & Public Services 1683 1939 256 12638 14006 1368

BC Rest of Canada

-200

0

200

400

600

800

1000

1200

1400

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

New Entrants Net In-Migration Net Other Mobility


21 September 2014

Figure 16 Sources of Total Supply Change for Heavy Equipment Operators, Rest of Canada

The number of new entrants is a more stable source of supply and easier to target. Table 6 shows the expected number of new entrants for operators across the county. As one would expect they are found in the larger provinces and with relatively large demands for operators. Ontario, Quebec, and British Columbia stand out as locations from which new entrants could be obtained.

Table 6 New Entrants, Heavy Equipment Operators

Another source of labour force supply is one that results because labour force supply and demand do not always match. It is the amount of excess supply in an occupation, which is shown in Table 7 below for all provinces and was displayed in Table 2 for Alberta. Excess supply is the difference between unemployment and normal unemployment. This unemployment is a measure of cyclical unemployment that arises as the economy goes through economic cycles. Cells in the table where excess supply is positive are highlighted. The provinces where positive excess supply is observed are places from which organizations could attract operators. There is significant excess supply to 2015 reflecting slower growth in the economy. Nevertheless, there is some excess supply in most provinces that can be accessed in the medium term as well. This situation is particularly the case in Quebec where excess supply remains a strong positive over the forecast. While organizations can attempt to attract part of an occupation’s normal unemployment, it will likely be more difficult to do so than for those unemployed in excess of normal levels. This is particularly the case for seasonal occupations where workers are only temporarily without work and are expecting to start work again in the near future.

-2000

-1000

0

1000

2000

3000

4000

5000

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

New Entrants Net In-Migration Net Other Mobility

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Canada 2192 2199 2188 2178 2169 2167 2160 2139 2129 2138 2157

British Columbia 292 295 296 304 304 304 298 293 281 277 276

Alberta 462 480 479 477 477 485 495 504 512 524 537

Saskatchewan 147 142 145 142 139 142 140 138 134 134 136

Manitoba 93 92 93 96 99 97 95 94 95 93 92

Ontario 517 518 518 513 513 511 510 499 493 489 486

Quebec 431 424 414 408 405 400 396 391 394 400 409

New Brunswick 77 74 75 77 80 82 79 75 75 75 76

Nova Scotia 83 83 80 75 73 71 70 69 69 68 68


Newfoundland & Labrador 77 79 77 74 68 66 65 65 65 65 65


September 2014 22

Table 7

Excess Labour Supply, Heavy Equipment Operators

The information shown above suggests that there will be challenges in obtaining operators over the next few years for the oil sands organization in question as the major source of supply of workers will be from outside the province. This sourcing will be difficult in the medium term as other provinces also face this situation. In the long term sufficient workers will be available as expansion demand weakens across the country. It will be necessary to attract new entrants to the occupation in Alberta – and obtain some from other provinces through inter-provincial migration. The POMS assumes that an occupation will get its share of total new entrants to the provincial economy based on its share of employment in the provincial economy as a whole. This is only an assumption and it is important for organizations requiring operators to try their best to attract young people into the occupation. There are not enough young people coming into the labour force as a whole to meet expansion and replacement demand through new entrants to the occupation. This example illustrates the type of information produced by POMS that can serve as a starting point in an occupation analysis for organizations. It is available for 500 occupations in each province where those occupations exist. In the smaller provinces there are fewer occupations because of the size of the labour force and diversity of occupations.

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

Canada 621 190 194 -160 -441 -442 -388 -398 -173 -70 -18

British Columbia 178 62 0 -28 -115 22 36 9 65 46 23

Alberta -21 -97 4 -30 -90 -155 -202 -192 -133 -88 -45

Saskatchewan -61 12 17 12 20 -20 -1 1 37 20 12

Manitoba 38 30 14 -30 -32 -1 21 12 10 26 28

Ontario 179 37 -86 -220 -308 -320 -261 -189 -103 -29 2

Quebec 223 100 154 92 54 62 78 51 39 8 -8

New Brunswick 74 88 46 -11 -47 -61 -32 -25 -29 -19 -12

Nova Scotia 70 24 30 27 10 -6 -17 -22 -19 -14 -14


Newfoundland & Labrador -74 -80 1 19 56 29 -14 -48 -47 -30 -16


23 September 2014

Occupation Demand and Supply Models There is a model for determining occupation employment and one that determines occupation demand, occupation supply, as measured by the labour force, and the components of labour force change. Each of these models takes its inputs from a macroeconomic model as well as other assumptions associated with the particular model and variables. The discussion below describes the structure of the model using both equations and flow charts where possible. It is included for those persons interested in knowing how the modelling system works.

Employment The model for employment is based on the widely used employment share approach. The employment share approach estimates an occupation’s employment as a share of its employment in the various industries in the economy. In the case of the manufacturing industry, for example, the equation for an occupation, J, is of the form:

OCCE(J,MA) = ROCCE(J,MA)*EMA where OCCE(J,MA) is employment for the occupation J in manufacturing, ROCCE(J,MA) is the share of employment in manufacturing and EMA is total employment in manufacturing. For example, if chemical engineers’ share of manufacturing industry employment is 5 percent and total employment in the industry is forecast to be 1000, then the forecast of chemical engineers employed in the manufacturing industry is 50. To determine total employment for an occupation, the model sums the individual industry calculations across all industries in which the occupation is employed. The employment shares are variables in the model and can be changed over time. In some cases they are modelled based on relevant determinants of employment. For example, in the case of the elementary and secondary school education services industry, the shares of elementary and kindergarten teachers and secondary school teachers are modelled separately based on the distribution of the respective school-age population. As the population associated with elementary and kindergarten teachers of both categories increases so does the share of employment and vice versa. Movements in the employment shares are allowed to change relative to their long run values in response to the economic cycle. Over the cycle it is assumed that overhead occupations such as managers and supervisors exhibit a much smaller cycle than non-overhead occupations. In this case, the shares of overhead occupations in an industry’s total employment increase relative to their normal values in economic slowdowns and fall relative to these values in economic recoveries. The opposite movement is assumed for non-overhead occupations. When employment growth stabilizes the shares return to the long run values specified for them. To provide accurate information on occupations at a 4-digit NOC level it is important that one uses relatively detailed industry employment categories. The POMS contains 70 industry categories including total employment. These industries are shown in Table 2.

Labour Force Demand and Supply The model that determines labour force demand and labour force supply along with its components of change starts first by determining an occupation’s labour force using what might be called a labour force share model. It then calculates the components of the change in the resulting labour force for the occupation. The labour force model is based on the assumption that the labour force for an occupation in the long run will be determined by the demand for the occupation. In the short run, there will be deviations between the growth rate in supply and demand because of mobility restrictions such as the time required to attract workers to the occupation either from other occupations or through the school system, as well as to get them to move from a different geographic area.


September 2014 24

Table 2

Industry Categories

A further restriction placed on the adjustment of the labour force for an occupation is that employers requiring the occupation must compete with each other and with other occupations to gain a share of the economy’s overall labour force. The latter variable is produced in the macroeconomic models. The ability to compete for additional supply of an occupation is driven by the occupation’s share of overall labour requirements in the economy. The higher its requirements relative to other occupations the larger is its share of the economy’s overall labour force The measure of demand for an occupation is the number of employees required, employment, plus the number of persons “normally” unemployed. The latter measure refers to the sum of frictional, seasonal, and structurally unemployed persons as described above. For an occupational labour market to function properly it normally requires some unemployment to allow people to change jobs and meet seasonal peaks in employment. In addition, it is normally the case that the skills of some of the unemployed do not match those required by employers, as is indicated by positive vacancy rates for occupations while unemployment rates are greater than zero. The measure of labour force occupation demand for occupation J, LFOD(J), is computed using employment and the occupation’s normal rate of unemployment, as shown below:

LFOD(J)=E(J)/(1-UNR(J)/100) E(J) is employment for occupation J and UNR(J) is the normal unemployment rate for occupation J. The labour force supply equation for an occupation focuses on what might be called the possible labour force. This variable is computed as a share of the total labour force in the economy, where the share is the ratio of the occupation’s labour force demand to labour force demand for all occupations in the economy as a whole:

PLFOD(I) = LFOD(I)/LFOD * LF

Forestry and logging Support activities for transportation

Fishing hunting and trapping Postal service plus Couriers and messengers

Support activities for agriculture and forestry Warehousing and storage

Farms Other Transportation

Oil and gas extraction Publishing industries

Support activities for mining and oil and gas extraction Motion Picture and Video Industries

Mining Broadcasting (except Internet)

Utilities Telecommunications

Residental Building Construction Other Information & Culture

Non-residential Building Construction Insurance carriers and related activities

Heavy and Civil Engineering Construction Finance

Trade contracting Real Estate Rental and Leasing

Food manufacturing Architectural, Engineering and Related Services

Wood product manufacturing Computer Systems Design and Related Services

Paper manufacturing Management, Scientific and Technical Consulting Services

Petroleum and coal products manufacturing Other Professional, Scientific and Technical Services

Chemical manufacturing Management of Companies and Enterprises & Administrative and Support

Non-metallic mineral product manufacturing Elementary and secondary schools

Primary metal manufacturing Community colleges and C.E.G.E.P.s

Fabricated metal product manufacturing Universities

Machinery manufacturing Other Education

Computer and electronic product manufacturing Ambulatory health care services

Electrical equipment appliance and component manufacturing Hospitals

Transportation equipment manufacturing Nursing and residential care facilities

Furniture and related product manufacturing Social assistance

Other Manufacturing Performing arts spectator sports and related industries

Wholesale Trade Amusement gambling and recreation industries

Motor vehicle and parts dealers Other Amusement and recreation

Health and personal care stores Accommodation services

Other Retail Trade Food services and drinking places

Air transportation Other Private Services

Rail transportation Federal government public administration

Water transportation Provincial and territorial public administration

Truck transportation Local municipal and regional public administration


25 September 2014

LFOD is total labour force demand and LF is total labour force in the economy as a whole. LF is normally not equal to LFOD as the latter represents “desired” labour force, not the actual labour force. The model assumes that it takes time to achieve the possible requirements because of the restrictions mentioned above. This assumption is implemented by using a partial adjustment model:

LFOD(J,T) - LFOD(J,T-1) = ADJ*(PLFOD(J,T) - LFOD(J,T-1)) where ADJ is the adjustment coefficient (takes values between 0 and 1), T refers to the current year and T-1 to the previous year. In this equation the labour force adjusts over time to achieve the possible requirements where the speed of adjustment is measured by the size of the adjustment coefficient. A value of 0.5 for the coefficient implies that almost all of the adjustment would take place in two years The basic structure of the approach used to estimate occupation demand and supply (stocks) described above is shown in Figure 8. The variable T in the figure represents a particular year. The first step in the approach is to determine occupation employment using assumptions about industry employment from the macroeconomic model and assumptions regarding the shares of occupation employment by industry. Employment is then adjusted to account for required normal unemployment to obtain occupation demand. Possible occupation requirements are then obtained from the labour force for the economy as a whole. These requirements are used in conjunction with last year’s supply and the adjustment coefficients to determine occupation supply for the current year.

Figure 8 Occupation Demand and Supply Determination

Economic Model

Industry Employment (T)

Occupation Supply (T)

Occupation Requirements (T)

Normal Unemployment Rates (T)

Total Labour Force (T)

Assumptions

Occupation Industry Shares (T)

Occupation Supply (T-1)

Supply Adjustment Coefficients (T)

Occupation Employment (T)

Possible Requirements (T)

Total Occupation Requirements (T)


September 2014 26

Labour Force Change The next part of the occupation supply model determines the sources of change in the labour force from one year to the next. The labour force change for an occupation is divided into the following components:

Deaths Retirements New Entrants Net In-Mobility

o Net In-Migration (net international and net interprovincial) o Other net in-mobility (inter-occupation mobility, cyclical factors)

As an equation this change for occupation I is expressed as:

LF(J,T) – LF(J,T-1) = LFNE(J,T) – LFDEATHS(J,T) – LFRETIRE(J,T) + NINMOB(J,T) Where LFNE(J,T) is new entrants, LFDEATHS(J,T) is deaths, LFRETIRE(J,T) is retirements and NINMOB(J,T) is net in-mobility. To compute these components of change it is necessary to consider population by age groups. The model uses 10-year age groups starting with the group ages 15 to 24 and ending with age 65 and over.

Deaths and Retirements The number of deaths for an occupation in an age group, J, is modelled as the product of an age-specific death rate, RLFDEATHS and the number of people in the labour force of that age group:

LFDEATHS(I,J,T) = RLFDEATHS((I,J,,T) * LF((I,J,T-1) Retirements are modelled as a product of a retirement rate for an age group and the number of people in the labour force of that age:

LFRETIRE(I) = RLFRETIRE((I,J,T) * LF((I,J,T-1) The age groups for retirements start at the group 45 to 54. The determination of deaths and retirements are shown in Figure 9. T refers to a year and I to an age group.

New Entrants The number of new entrants to the labour force refers to persons entering the labour force from the population age groups 15 to 24 and 25 to 34. This concept is meant to refer to persons that enter the labour force either for the first time after receiving an education or on a more permanent basis as described above. It does not include migrants or occupations such as managers that require related labour market experience.

The first step in estimating new entrants is to estimate the number of new entrants for the economy as a whole – see Figure 10. This is accomplished using labour force participation rates for the two age groups and the number of persons in the age groups. The model then shares the new entrants into an occupation out of the total new entrants in the age groups. The recent ratio of employment in an occupation to total employment in the economy as a whole is the share employed. The assumption in this case is that if the share of an occupation in the economy is rising then so will its share of new entrants and vice-versa. The equation for new entrants for occupation J, LFNE(J) is of the form:

LFNE(J) = RLFNE(J) * LFNE where RLFNE(J) is the recent demand share and LFNE is total new entrants.


27 September 2014

Figure 9 Deaths and Retirements Determination

The new entrants for the occupation age groups are computed as the product of the normal proportion of new entrants in the age group multiplied by the occupation new entrants. The latter proportion reflects the age at which the occupation enters the labour force. The proportion of physicians in the 15 to 24 age group, for example, is much lower than it is for the 25 to 34 age group. This situation is a result of the time required to obtain the educational requirements for a physician.

Figure 10 New Entrants Determination

Occupation Supply (T-1,I)

Assumptions

Retirement Rates (T,I)

Occupation Retirements (T,I)

Death Rates (T,I)Occupation Deaths (T,I)

Multiply

Multiply

I – Age Group

T - Year

Occupation Demand Share (T)

Total New Entrants (T)

Economic Model

Average Demand for all Occupations

(T)

Average Occupation Demand (T)

Occupation New Entrants (T)

Multiply

Divide

Age Group Share Assumption (T,I)

Occupation New Entrants (T,I)

Multiply

I – Age Group

T - Year


September 2014 28

Net In-Mobility (Net In-Migration and Other Net In-Mobility) The last two components of labour force change measure the change in the labour force that is required because withdrawals from the labour force – deaths and retirements – are not equal to new entrants to the occupation’s labour force. This difference must be made up from inter-occupational mobility as well as net in-migration. There is also a cyclical component of the labour force which shows withdrawals from the labour force increase as employment falls and vice versa (discouraged worker effect) as described above. These two components as a whole, net in-mobility, is computed as a residual from labour force change given new entrants, deaths, and retirements:

NINMOB(J,T) = (LF(J,T)-LF(J,T-1)) – LFNE(J,T) + LFDEATHS(J,T) + LFRETIRE(J,T) They are then separated into a net in-migration component and the other net in-mobility component. The net in-migration component is computed from the labour force net in-migration determined in the macroeconomic models – see Figure 11. The net in-migration component for the occupation is shared out of total labour force net in-migration using net in-mobility requirements. In this case, it is the share of occupation net in-mobility requirements of total net in-mobility requirements that determines net in-migration associated with the occupation. The change in an occupation’s labour force through net in-migration, LFNM(J), can be represented as:

LFNM(J) = NINMOB(J)/NINMOB*LFNM where NINMOB is total net in-mobility and LFNM is total net in-migration. The labour force net in-migration is determined using total net in-migration by age and sex and age and sex specific labour force participation rates. Total net in-migration is determined outside the model. Because net in-mobility is a flow and can take on the value zero, an alternative measure is used to allocate net migration to occupations when total net in-mobility approaches zero. This measure is the occupation’s recent share of the labour force for the occupations as a whole.

Figure 11 Determining Net In-Migration

Multiply

Occupation In-Migration (T)

Occupation Net In-Mobility Ratio (T)

Total Labour Force In-Migration (T)

Economic Model

Net In-Mobility for all Occupations (T)

Occupation Net In-Mobility (T)

Divide

Age Group Share Assumption (T,I)

Occupation In-Migration (T,I)

Multiply

I – Age Group

T - Year


29 September 2014

Net in-migration for an occupation is split into net international in-migration and net interprovincial in-migration. Net international

in-migration for an occupation is shared out of total net international in-migration based on its share of net in-migration. Net

interprovincial in-migration for an occupation is computed as a residual from its net in-migration and net international in-

migration.

The other net in-mobility component, ONINMOB(J), is then determined as the difference between the occupation’s residual change and the change due to net in-migration:

ONINMOB(J) = NINMOB(J) – LFNM(J) The net in-mobility values are allocated to the age groups in the model using age shares. The net migration shares are based on the shares found in the migration data but are adjusted to reflect different age distributions across occupations. For other net in-mobility they reflect the existing shares of the age groups in the labour force as a whole, but are adjusted to reflect different age distributions across occupations. The shares employed differ across experienced occupation such as managers and occupations that show significant differences in their age structures from the occupations as a whole. Managers, for example, are generally older. Their 15 to 24 and 25 to 34 age group shares, for example, are expected to differ from that for new entrants. Other occupations such as sales clerks that require relatively fewer skills are relatively younger than are other occupations. They tend to move on to other occupations as they age. As a result, fewer of them are obtained from the older age groups for net in-mobility. Occupations that require significant experience such as judges are found in older age groups. Adjustments to labour force age shares obtained for the labour force as a whole are made for these occupations.

Documents

Occupation modelling system - C4SEOCCUPATION MODELLING SYSTEM September 2014 2 Introduction This document describes the Canadian provincial occupational modelling system (POMS). POMS