Population Projections 2006 to 2031- Lambton

Page |1

COUNTY OF LAMBTON

Population Projections Census Years 2011 to 2031 An Update to Population: Summary Trends & Projections, Census Years 2006 to 2031, a Report Adopted September 1, 2010 by County of Lambton Council

Prepared by: Planning & Development Services Department 7/21/2010

Page |2

Population Projections 2006 to 2031- Lambton

Table of Contents Purpose ........................................................................................................................... 4 Executive Summary ........................................................................................................ 5 Introductory Issues .......................................................................................................... 7 Geographic Divisions ................................................................................................... 7 Small Populations ........................................................................................................ 7 Recalculated Projections ............................................................................................. 7 2011 Census Results and Ongoing Trends..................................................................... 9 Age Distribution ........................................................................................................... 9 Total Population ........................................................................................................... 9 Historic Growth Trends .............................................................................................. 10 Age-Specific Growth Rates........................................................................................ 11 Birth Rates ................................................................................................................. 13 The Projections’ Methodology ....................................................................................... 14 Theory and Assumptions ........................................................................................... 14 Projecting Future Population ..................................................................................... 15 Birth Rates ................................................................................................................. 15 Calculating Historic Age Specific Growth Rates ........................................................ 15 Projecting Future Age Specific Growth Rates ............................................................ 16 2006 Projection Methodology ................................................................................. 16 Refined Projection Methodology in this Report ...................................................... 17 Summary of Projection Methodology Changes .......................................................... 20 Population Projections................................................................................................... 21 Results....................................................................................................................... 22 Future Age Distribution and Explaining Declines ....................................................... 23

Population Projections 2006 to 2031- Lambton

Page |3

Sensitivity .................................................................................................................. 24 Monitoring ..................................................................................................................... 24 Adoption for Planning Purposes .................................................................................... 25 Recommendation .......................................................................................................... 25 Appendix A .................................................................................................................... 26 Projected 2031 Populations in Comparison with 2006 Projections ........................ 26 Projected 2031 Populations in Comparison with 2006 Projections ........................ 27 Appendix B .................................................................................................................... 28 Projected Age Specific Growth Rates .................................................................... 28 Projected Age Specific Growth Rates .................................................................... 29 Appendix C .................................................................................................................... 30 Actual 2011 Age-Specific Growth Rates in Comparison with 2006 Projections ..... 30 Actual 2011 Age-Specific Growth Rates in Comparison with 2006 Projections ..... 31

Page |4

Population Projections 2006 to 2031- Lambton

Purpose On September 1, 2010, County Council adopted the population projections contained in the report: Population: Summary Trends & Projections, Census Years 2006 to 2031 referred to in this report as "the 2006 projections". The purpose of the present report is to update the 2006 projections with 2011 census data. The first goal of this report is to describe the results of the 2011 census, particularly in comparison with historic trends and 2006 projections. It is not our intent to reiterate all of the content of the 2006 projections report, especially municipality by municipality. This is because the bigger trends continue in most cases. This report focuses on the 2011 census results that differed from previous trends and expectations. The second goal of the report is to outline the theory, assumptions, and mechanics of the projection methodology and the minor changes used by the projections contained in this report as compared to the 2006 projections. Fundamentally, the methodology has not changed from the 2006 projections. The changes and the detailed reasons for the changes are contained throughout this report, especially the Projection Methodology section. A short summary of all changes in methodology is contained towards the end of this report. The third goal of this report is to provide new projections for 2031 populations taking into account the results of the 2011 census. This includes a review of the methodology for calculating historic age-specific growth rates, projecting future age-specific growth rates, and applying age-specific growth rates to project future population and age distribution. The methodology's assumptions, strengths, and weaknesses are also discussed.

Population Projections 2006 to 2031- Lambton

Page |5

Executive Summary The projections predict the future age distribution and population that will be realized if growth trends continue at the rates experienced in the three most recent census periods. As was noted in the 2006 projections, Lambton County’s age distribution and growth trends in key age groups are not unlike most other largely rural counties. Therefore, the long-term projections for Lambton predicted by this report are not unique in Ontario. The general assumption of this project is that today’s age distribution has some influence on future total population. This is because growth trends and the ability to produce new population are assumed to differ from one age group to another. In Lambton, like all of Ontario, the baby-boomers make up a larger proportion the population. As these age groups enter the ages of higher mortality rates, they will experience large population losses. As the tail-end of the baby-boomers has left the child-bearing ages, the number of children in the population has declined significantly in recent decades. High rates of losses in young adults have been off-set by a large population of children to start with, such that the actual population of young adults has held somewhat steady in recent decades. With the reduced number of children now, the young adult population is likely to begin to decline if losses in young adults continue at historic levels. On the whole, the 2006 projections proved to predict 2011 census results with satisfactory accuracy. Most 2011 populations were within the expectations predicted by the various growth scenarios of the 2006 projections. Growth trends by age group, in particular, matched quite well with predicted percentages of growth. 2011 census results that fell outside of expectations include Sarnia, Lambton Shores, and to a lesser extent, Point Edward. Whereas Sarnia had been losing young families, it experienced greater growth (or lesser decline) in all ages from 0 to 39 than it had in any of the last three census periods. This resulted in unanticipated growth of 1.3%. 2011 census results raise projections considerably and suggest some potential for growth in the long-term. However, even with the influence of 2011 data, decline is still expected with an aging of the population. Lambton Shores experienced a large population gain in 2006 only to see a loss of these gains in 2011. Growth rates were below historic rates in all age groups. Instead of strong potential for gains, the long-term projection is decline with possibility of regaining 2011’s losses. As described in the report, there is suspicion of a quirk in the census data. The projections in this report have used the same fundamental assumptions as the 2006 projections but have made some modifications to the methodology that are based on more accurate assumptions and should improve accuracy. While results such as

Page |6

Population Projections 2006 to 2031- Lambton

those of Sarnia and Lambton Shores could not have been predicted from past trends, the changes would help to anticipate such events to some extent. Like the previous projections based on 2006 and earlier census data, the long-term expectation in Lambton County is population decline. A note of caution should be given about the projections. The methodology does not accommodate major changes in growth rates in the short term such as the establishment of a major employer. It may also be that there are natural checks and balances in place that would cause significant changes in some age group’s growth rates to compensate for losses. Finally, the projections do not show differences in growth trends for communities within municipalities. For example, it is expected that Corunna is experiencing some of the best growth in the County despite St. Clair showing slight decline. The projections do show however, that there are “macro-trends” that, along with the current age distribution, are leading towards over-all population decline. Growth in key age groups would have to be substantially different to reverse these trends. It is possible that these macro-trends could change. A recession in the economy or a more even age distribution (rather than a lack of 20-29 year olds) could result in fewer job opportunities in larger urban areas and lead to more youth remaining in Lambton. Staff may have experienced some of this in this last census period. Significant changes in the over-all age distribution could also lead to different trends in age-specific growth rates. Sensitivity analysis from the 2006 projections found that the most long-term benefits to over-all population would come by improving immigration rates in the 0-9 and 30-39 age groups, increased birth rates and reducing emigration rates in young adults. Improving growth rates in the retiree age groups would not have the same benefits. It should be noted that population decline does not necessarily mean a decline in development activity. Future work will be needed to translate these projections into housing demand, but initial analysis suggests that ongoing declines in average household sizes will translate into continuing demand for new dwelling unit construction.

Population Projections 2006 to 2031- Lambton

Page |7

Introductory Issues Geographic Divisions Both the previous and present projections have produced projections for the County as a whole and each lower-tier municipality. In order to look at differences in trends between urban and rural areas, the previous projections produced projections for a geographic entity staff referred to as Rural Lambton. This area consists of the entire County except Sarnia, Point Edward, and Lambton Shores. Although, not a strictly urban area like Sarnia and Point Edward, Lambton Shores' growth trends were unique enough that it was also excluded from Rural Lambton for projection purposes. With the present projections, staff has also found it useful to produce projections for a new geographic entity denoted as Under 4,000 Rural. This entity consists of the rural municipalities in Lambton with a population under 4,000 (Dawn-Euphemia, BrookeAlvinston, Enniskillen, Oil Springs, and Warwick). Staff has used this entity for moderating the rates of smaller rural populations.

Small Populations Municipalities with small populations present unique challenges for projecting future growth trends. With small populations, the populations of individual age groups can be small enough that their accuracy is suspect. Rather than reflecting actual population trends in the municipality, historic growth rates may actually reflect an individual's onetime choice or could even be the results of the way the census rounds data. This is reflected in the sometimes wide variation in growth rates from year to year in municipalities with smaller populations. The smaller the population, the more susceptible growth trends are to the influence of anomalies. In order to reduce the potential effects of anomalies, the projections moderate the growth trends of rural municipalities of smaller population against a larger geographic area with what are assumed to be similar growth trends. The previous projections moderated growth rates against those of Rural Lambton. The present projections, however, use the Under 4,000 Rural trends for moderation. The previous projections moderated Point Edward's trends against those of the County, however the present projections do not moderate Point Edward's rates against any other.

Recalculated Projections First Nations are not required by law to respond to the census survey. Since the 2006 projections were produced staff has become aware of the reduced data quality and potential for error introduced by fluctuating response rates. This means that what appear as population fluctuations may in fact reflect varying response rates. In particular, Walpole Island's response rate was low enough that it was completely excluded from the 2011 census.

Page |8

Population Projections 2006 to 2031- Lambton

Walpole's exclusion from the 2011 census impacts the County and Rural Lambton populations and influences all of the moderated growth rates used in the 2006 projections. This makes it impossible to compare 2011 census data to many of the 2006 projections. For these reasons, it was necessary to completely recalculate the 2006 projections using the same methodology, but excluding the Census Tracts classified as "Indian Reserves" (Kettle Point, Sarnia 45 (Aamjiwnaang), and Walpole Island) from the County population. Throughout this report therefore, references to the "2006 projections" are not actually referring to the projections adopted by Council, but rather the projections that were reproduced under the same methodology but excluding the census areas defined as Indian Reservations. The present projections and all further updates are based on a County population that excludes the Indian Reservations. This is necessary to remove the influence of varying response rates and to ensure that projections are comparable from census to census.

Figures 1 and 2 show the difference between the original 2006 projections and those that exclude Indian Reservations. The predicted growth rates in Figure 1b, which exclude Indian Reserves, closely match actual 2011 rates in comparison to those rates including them in Figure 1a. Rural Lambton is used as an example here, but the same is true for the County as a whole.

Population Projections 2006 to 2031- Lambton

Page |9

2011 Census Results and Ongoing Trends Age Distribution Figures 3 and 4 show the 2011 age distribution for the County and Sarnia in comparison with previous census years. They show what has become known as the typical “boom, bust, and echo” age distribution. The aging of baby boomers is quite apparent, and explains the declining numbers of middle-aged population. Given the high percentage losses in young adults, it is notable that the actual number of people in those age groups has stayed comparatively constant over the last twenty years. This is apparently because the starting numbers of children have been sufficiently large to offset large losses when they become young adults. The ability for this trend to continue is questionable however given the way the child population has declined over the last 20 years. A significant decline in child-aged population is apparent compared to 1991. This is a result of a general and steady decline in birth rates from 1991 to 2006 and also less population in child-bearing years compared to 1991 when the tail-end of the babyboomers cohort was still having children. Fig 3

Fig 4

Total Population The 2006 projections included a total of five growth scenarios. Figures 5, 6 and 7 show the actual 2011 Census populations alongside two of these scenarios: the "predicted" and "target" growth scenarios. The actual 2011 populations are generally within the expectations of the 2006 projections. Those 2011 populations that fell outside of anticipated ranges are Sarnia, Lambton Shores, and Point Edward. Sarnia's population grew by approximately 935 or 1.3%, whereas past growth trends were predicting decline. The most optimistic of the 2006 projection scenarios was growth of only 0.3%.

P a g e | 10

Population Projections 2006 to 2031- Lambton

Lambton Shores' population declined by approximately 505 or 4.5%. It had experienced very strong growth in recent censuses. The "predicted" scenario projected 2.9% growth. The least optimistic projection was only a 2.8% decline.

Point Edward's growth of 0.5% or about 10 individuals was also outside of 2006 projections, which predicted losses of 15 or more individuals given recent trends. This variation is not surprising given the small population size. All of the rural municipalities with small populations had their rates moderated against those of Rural Lambton. Although their 2011 populations are all within the range of 2006 projection expectations, it is notable that they are each a little below the "predicted" 2006 projections.

Historic Growth Trends This report does not itemize every growth rate for every age, year, and municipality, but age-specific, demographic trends can be seen for each municipality in the graphs in Appendix B to this report. They show "reference", "high", and "low" projected growth rates for each municipality. These terms and their role in producing projections are explained in greater detail in Projection Methodology section. The "reference" rates are an average of the historic growth rates of the last three census periods, and the "high"

Population Projections 2006 to 2031- Lambton

P a g e | 11

and "low" growth rates provide an indication of which age groups have the greatest fluctuation in rates from census to census. Lambton as a whole is experiencing modest growth in working age population, in children (of the working age population), and in younger retirees. It continues, however, to lose youth and young adults in quite high percentages. With some variation in the age ranges and magnitudes at which these trends are being experienced, these trends generally hold true for local municipalities. A more detailed description of growth rates and trends being experienced in each municipality is included in the 2006 projection report. Those 2011 census results that appear as deviations from the 2006 trends are described in this report.

Age-Specific Growth Rates Appendix C contains figures showing the actual 2011 age-specific growth rates in comparison with the 2006 projected rates. The actual 2011 age-specific growth rates for the County and local municipalities correspond very well with the 2006 projections. Figure 2 (on page 8) and Figure 8, for example, show high correspondence between projected and actual 2011 growth rates for Rural Lambton and the County as a whole. The reason for Sarnia's 1.3% population growth is apparent from Figure 10. Sarnia's growth trends were accurately predicted, but not their magnitudes. No age group's growth rate was far off from what was predicted, but there was a small increase over average historic rates for all ages 0 to 39. Lambton Shores' growth rates (Figure 9), on the other hand, were below predictions in every age group. Their trend was generally as predicted, however, their

P a g e | 12

Population Projections 2006 to 2031- Lambton

magnitude was well below historic levels. Dwelling count data suggests the declines in Lambton Shores' population are tied to a significant shift in homes being converted from permanent to seasonal. Staff suspect this may reflect some kind of data quirk or change in how data was collected or categorized by Statistics Canada in either 2006 or 2011. The 2011 declines could actually be a correction of artificially high population gains in 2006. The projected growth rates for St. Clair, Plympton-Wyoming, and Petrolia had extremely good fit with actual 2011 growth rates. Warwick experienced greater variability in growth rates than predicted. The degree of population decline in Warwick, although not unexpected, was largely influenced by lower than predicted growth rates in ages 20-34. A high variability in growth rates due to the high potential for anomalous rates is expected for municipalities with populations under 3,500. This was the case in 2011. Despite this, trends in small municipalities were predicted well with few exceptions. Oil Springs should be noted as a special case. Because of its population size, agespecific growth rates fluctuate significantly such that there were no identifiable trends in growth expected or observed and there will be a greater degree of uncertainty in any projections for this municipality compared to others. Having moderated small rural populations' rates with those of Rural Lambton seems to have been effective in reducing anomalous influences. Rural Lambton includes some larger populations and urban areas however, and their inclusion may have introduced some inappropriate growth trends into the predictions for the smaller populations. For example, Enniskillen in 2011 had much greater than predicted losses in ages 25 through 34. Dawn-Euphemia had much greater than predicted losses in ages 15 through 34. Moderating their rates hid these trends. As noted, the actual 2011 populations for those municipalities whose growth trends were blended with Rural Lambton were slightly, but consistently below the "predicted" 2006 projections. Moderating the rates with Rural Lambton may have not only eliminated some anomalies but also some real trends in these municipalities and seems to have artificially raised the 2006 projections. Point Edward's slight growth appears to be the result of growth among seniors at higher than historic rates. The population changes in Sarnia and Lambton Shores stand out as the most significant changes in previous trends. Staff do not wish to suggest that there is anything fundamentally wrong with the projections or that these changes should have been foreseen. However, staff have made a number of refinements in the methodology used in producing the current projections. These are discussed in the Projections Methodology section of this report. Staff believe these changes will help address some of the issues noted above, are based on more accurate assumptions, and will further improve accuracy.

Population Projections 2006 to 2031- Lambton

P a g e | 13

Birth Rates Birth rates have generally been declining consistently the past few decades and the projections assume they have leveled out. However, in 2011 the ratio of children aged 0-4 to adults in child-bearing years actually increased in many municipalities to levels similar or slightly higher than 2001. The projections currently use the child-to-adult ratios of the most recent Census period and staff recommend no change.

P a g e | 14

Population Projections 2006 to 2031- Lambton

The Projections’ Methodology Theory and Assumptions Having more or less population in a particular age group may or may not be an indication of whether the municipality is attracting or losing people in that age group. The future population of a particular age group does not depend on the size of the population currently in that age group. It depends on the populations of the younger, up-and-coming age groups and the amount by which those age groups will grow or decline before they reach the particular age in question. The first fundamental assumption of these projections is that the interaction between the current age distribution of a municipality and the rates of population change in individual age cohorts is the most significant determinant in the municipality's projected population. The second fundamental assumption of these projections is that the growth rate that a particular age group in a particular municipality will experience over the course of a census period will be similar to the rates of growth (or decline) that other groups experienced in previous census periods when they were that age. The realization of the projections depends more than anything else on the ability to predict these age-specific growth rates for key age groups. For most municipalities, historic growth rates do show identifiable trends from one age group to the next and suggest that the above assumption is reasonable and that it is possible to expect a growth rate to fall within a certain range of values for a given age group in a given municipality. Our 2006 projections were based on these assumptions. On the whole, they predicted actual 2011 growth trends and age-specific growth rates quite well. There were relatively few events that were not anticipated from past trends. Given the nature of the methodology, the immense number of individual growth rates involved in the calculations, and the laws of averages, a number of future growth rates will fall out of historic bounds. A difficulty of this methodology is in not knowing which are anomalies or fluctuations around a norm and which are the beginnings of actual changes in trends. An implication of the methodology’s assumptions is these rates will be treated as fluctuations or anomalies. Historic rates suggest that this is the case in the majority and a reasonable assumption. To treat every anomaly as a change in trends would quickly make the projections meaningless. These outlying rates will be factored into and influence the averages and the allowed variation, however they will have to prove themselves as such before being treated as changes in trends. Where new trends have shown themselves to be more than just anomalies or fluctuations, the methodology is self-correcting over time. The methodology is best suited to predict the long-term implications of current trends, but does not change quickly in the short-term if there are sudden departures from historic trends that are more than just anomalies.

Population Projections 2006 to 2031- Lambton

P a g e | 15

Projecting Future Population This projection methodology projects the total population for the next census by projecting the size of the individual five-year age cohorts for each municipality and adding them up. Most age cohorts’ populations are projected by applying an agespecific growth rate to the age group of one census that will become the age group being projected in the census in question. The first census period in the projections (i.e. 2016) will be based on the actual data of the most recent census period (i.e. 2011). The projections for the first census are in turn used as the basis on which the subsequent census’ projections are built and so on. The 0-4 age group is projected as proportionate to the child-bearing–aged population of the census being projected. The most difficult part of the projections then is projecting age-specific growth rates and birth rates. Projections for the County, Rural Lambton, and the Under 4,000 Rural municipalities are estimated by this same process. That is, they are not estimated by adding up the projected populations of their constituent municipalities. In fact, because of varying growth rates, the projections for individual municipalities will not add up to those of the whole.

Birth Rates The 0-4 population cannot be predicted in the same way as other age groups. Staff cannot base it on a percentage growth of a previous population, because it consists of people who were not yet born at the time of the previous census. The projections predict a census's 0-4 population from the projected population in childbearing age. It assumes the proportion of population aged 0-4 to population of childbearing age in future censuses will be the same as in the most recent census. The calculation takes into account that people in those age groups are not the same age throughout the census period, as this has implications for the number of people in childbearing years over the course of the census period. With the 2006 projections, this proportion of children-to-child-bearing age population was presented in the form of a live birth rate - number of births per thousand women. With the present projections staff have changed this to be presented as a ratio of children to child-bearing-aged population, which makes it simpler to understand how the 0-4 population is predicted. This is only a change in how the birth rate is presented. It does not change the mechanics of the projections of the predicted size of the 0-4 population in any way.

Calculating Historic Age Specific Growth Rates A first step in the projections is to calculate actual, historic, "age-specific growth rates" for each age group in each municipality in each of the three most recent census periods. In understanding the mechanics of calculating an age specific growth rate it is important to remember that the individuals making up a five-year age cohort in one census are the same individuals as five years ago, except they have aged by five years and are now in

P a g e | 16

Population Projections 2006 to 2031- Lambton

the next-oldest age cohort. Their population will also have changed due to new members through immigration and lost members by death or emigration. An age-specific growth rate for a particular age group, in a particular census period is the difference in the group's population in that census year and its population in the previous census year, divided by its population in the previous census year. For example, a 1996 age-specific growth rate for ages 30-34 would be calculated as follows:

A more detailed discussion of the calculation of age-specific growth rates is included the 2006 projections report.

Projecting Future Age Specific Growth Rates Staff feel that the assumptions behind the 2006 projections were fundamentally sound and the resultant projections proved to have predicted growth trends and populations with reasonable accuracy when compared with actual 2011 census data. Still, staff have made some refinements to the projection methodology that staff feel are based on more accurate assumptions. A discussion of the 2006 projection methodology and its assumptions is included in order to explain the differences in the methodology used in this report’s projections. The projections in this report are based on the three census periods from 1996 to 2011. The previous projections were based on the census periods from 1991 to 2006. 2006 Projection Methodology The 2006 projections included a total of five different growth scenarios with varying assumptions about the relationship of future and historic growth rates. These scenarios, their assumptions, strengths, and weaknesses are discussed in detail in the 2006 projection report. These growth scenarios were used to create “predicted” growth rates and “target” growth rates as summarized in Tables 4 and 5 of the 2006 projections report. Predicted Rates Predicted rates were based on the "weighted" scenario, an average of the rates of the last three census periods. As the name implies, it was not a straight average. More weight was placed on the more recent census periods. The most recent census period had twice as much influence on the predicted rate as the middle census period and three times as much weight as the most distant of the three census periods. For smaller municipalities with populations under 3,500, the predicted rate was found by further averaging the "weighted" rate of the municipality with that of Rural Lambton, or with that of the County in the case of Point Edward. These rates were called the "Rural Blend" and the "County Blend". Twice as much weight was placed on the rates of the larger area as on those of the municipality itself.

Population Projections 2006 to 2031- Lambton

P a g e | 17

An assumption of the predicted rates was that the most recent census periods have the most relevance for predicting future rates. In a number of municipalities, 2006 rates differed from previous rates considerably, and 2011 rates returned to the earlier rates. Putting more emphasis on 2006 data therefore caused 2011 populations to be predicted less accurately. 2011 census results suggest that an assumption that rates are more cyclical in nature would have provided an even greater level of accuracy in predicting 2011 populations. The process of blending the rates for smaller municipalities was intended to help eliminate the effects of anomalous rates that are more common among smaller populations. It also seems it had the effect of blending in trends from the larger rural municipalities that were not representative of the smaller municipalities. This caused slight decreases in the accuracy of projected populations for those municipalities. The effects of this assumption on the actual rates in comparison with 2006 projections can be seen in Appendix A. Target Rates Target rates for an individual age group were set according to the "Max" growth scenario. This rate assumed that the growth rate for each age group would match the maximum growth rates experienced in a municipality for that age group over the previous three census periods. A weakness of this scenario, already realized with the previous projections, was that this assumption provided no potential for growth rates to fall outside of the recent highs and lows. It was assumed that if an "abnormal" rate were to occur, it would average out with other groups or over the long-term and that it would not greatly impact short-term or long-term projections. In the case of Sarnia in 2011, rates exceeded historic highs marginally but consistently over a large number of age groups. This resulted in a higher than projected population in the short-term. The case was exactly the opposite situation for Lambton Shores. Overall, these events could not have been fully predicted from past trends and may or may not reflect long-term trends, but scenarios that would have allowed potential for rates to fall outside historic highs and lows would better anticipate such possibilities. The concern with the "Max" scenario not allowing rates to exceed past rates is especially a concern when one of the three years' growth rates is very low compared to the other two years. Conversely, in the situation where a "Max" rate is very high compared to the others, an opposite concern exists. In this case, the "Max" scenario is more likely to be promoting anomalous rates. Refined Projection Methodology in this Report The methodology used for the projections of this report produces a noticeable difference in appearance but is fundamentally the same as the 2006 projections. The current projections have "reference", "low", and "high" projections as opposed to the 2006 "predicted" and "target" projections. The calculated rates for each municipality are diagramed in Appendix B.

P a g e | 18

Population Projections 2006 to 2031- Lambton

Reference Scenario The new "reference" scenario is exactly the same as the 2006 "predicted" ("weighted") scenario, except that the age-specific growth rates are a straight average of the last three census periods’ growth rates. The 2006 "predicted" rates put more weight on more recent census periods. This change to a straight average is a change in assumption that rates are usually cyclical. In this way, "anomalous" rates will still influence the average but not be treated as a change in trend until they have held for two or three census periods. Staff anticipate this will make the projections less susceptible to the effects of anomalous years. Blended Reference Scenario The "reference" scenario growth rates of the County and municipalities with populations greater than 4,000 will be calculated strictly from their own historic growth rates. As with the 2006 projections, the "reference" rates for smaller rural municipalities are still moderated against the "reference" rates of a larger area. The entity with which they are blended however has changed from Rural Lambton to a new geographical entity we have called Under 4,000 Rural. This is a combination of the rural municipalities with populations under 4,000 – i.e., Dawn-Euphemia, Brooke-Alvinston, Enniskillen, Oil Springs, and Warwick. Staff have found its growth trend correspond better with those of the individual municipalities than do those of Rural Lambton. Blending the rates with this entity should have the benefit of reducing the effects of anomalous rates without introducing the differing trends of Plympton-Wyoming, Petrolia, and St Clair. Further, the blends are a straight average of the municipality's and the Under 4,000 Rural growth rates. Although this increases the risk of anomalous influences, its allows the trends of the local municipality to have more influence on the final projections. Another change to note is that the population size which defines whether a municipality's rates are based on its own historic growth rates or are blended has been changed from 3,500 to 4,000. This means the "reference" scenario for Warwick will be based on growth rates that have been blended with the Under 4,000 Rural rates. This is done because of the higher than expected variability in growth rates in the 2011 census, in particular the large losses of young adults. Point Edward is an urban municipality with a population below 4,000, and its "reference" rates have not been blended with any other area. As a municipality with a small population this will make it susceptible to anomalies. Staff have found, however, that Point Edward's historic growth trends are different from both the County and Sarnia such that continuing to blend Point Edward's rates is likely to produce more error than would the anomalies inherent in a small population. Low and High Scenarios The biggest change in methodology is the addition of a "low" growth scenario, and the use of a "high" growth scenario instead of the "target" or "max" growth scenario used in the 2006 projections.

Population Projections 2006 to 2031- Lambton

P a g e | 19

For any given age-specific growth rate, the "high" and "low" scenarios provide the possibility for the future rate to vary from the "reference" growth rate (the three year average) by a set amount. The amount of variation allowed is set by the range of rates experienced over the last three census periods. The amount of variation or range is equally distributed around the "reference" rate. Mechanically, the "high" and "low" are calculated as follows. First, the range is the difference between the maximum and minimum of the age-specific growth rates experienced in the last three census periods. For example, if the age-specific growth rates of the last three census periods are 7%, 1%, and 7%, the range is 6% and the average or "reference" rate is 5%. Second, half the range (3% in this example) is added to the "reference" to produce the "high" growth rate and half is subtracted from the "reference" to produce the "low" growth rate. In other words the projection would predict that the growth rate in that age group would be between a "high" of 8% and "low" of 2%, but most likely the "reference" rate of 5%. In the event that there is historically a large amount of variation in a particular agespecific growth rate from one census period to the next, the "high" and "low" growth rates will show the potential for much variation around the projected "reference" rate. In the event that historic rates have been very consistent, the "high" and "low" rates will predict that the actual growth rate will be very close to the projected "reference" rate. In the event that one of the three historic growth rates is substantially lower than the other two, the "high" scenario treats that lowest growth rate as an anomaly and will allow the possibility that the actual future growth rate could exceed historic rates. On the other side, if one of the three historic growth rates is substantially higher than the average, this approach would treat the historic high as an anomaly and allow the possibility that the future actual rate could be lower than historic rates. Although the resultant projections are not significantly different, staff feel that the use of the "high" and "low" growth scenario is built on more accuraqte assumptions, whereas the use of the "max" growth scenario as a target is somewhat arbitrary and subject to anomaly. On the one hand, the "max" growth scenario promoted what may be anomalously high rates. On the other, it did not allow rates to improve over previous census periods, which is a concern especially when one anomalously low rate has lowered the average. In the 2011 Census Results section of this report, Oil Springs is noted as an exception because of its population, high variability in growth rates, and lack of predictable trends. The general methodology of the “high” and “low” growth scenarios is to use an agespecific growth rate’s full historic range to predict its potential variability from the “reference” scenario. Because of the extreme variability in rates and to keep long-term projections reasonable however, the projections use only 60% of the range in predicting the “high” and “low” growth scenarios for Oil Springs’ age-specific growth rates.

P a g e | 20

Population Projections 2006 to 2031- Lambton

Summary of Projection Methodology Changes The following lists the various changes to the projection methodology noted throughout this report. 1. The County, Rural Lambton, Under 4,000 Rural populations presented in the report and used for producing projections excludes Indian Reserve populations. 2. The method for predicting the 0-4 population is given in the form of a ratio of children-to-child-bearing-aged populations, consistent with how it is actually predicted. 3. The projections are presented as “low”, “reference”, and “high” growth scenarios. 4. The reference projection is based on a straight average of the growth rates in the last three census periods. 5. The ‘high” and “low” projections provide equal potential for a future growth rate to vary from a predicted “reference” rate either above or below the rate. The difference between the “high” and “low” rates is equal to the range in rates experienced historically. This allows potential for growth rates to go beyond the bounds of historic rates. 6. Point Edward’s rates are not moderated against any other area. 7. Projections are produced for a geographic entity called Under 4,000 Rural which consists of the rural municipalities with a population under 4,000. 8. The growth rates of rural municipalities with smaller populations are moderated against the growth rates of the new Under 4,000 Rural entity rather than Rural Lambton. The municipality’s growth rates are given equal weight with those of Under 4,000 Rural. 9. The population below which a rural municipality’s growth rates are moderated against Under 4,000 Rural is 4,000 rather than 3,500 and now includes Warwick.

Population Projections 2006 to 2031- Lambton

P a g e | 21

Population Projections The projected populations under the “low”, “reference”, and “high” growth scenarios for each municipality are summarized in Table 1 and are visually displayed along with historic population growth back to 1991 in the graphs of Appendix A. Table 1: Population Projections 2011 Population

Growth Scenario

Lambton County

124,615

Low Reference High

119,417 122,263 125,109

113,952 119,387 124,963

108,060 115,891 124,122

101,349 111,367 122,162

Sarnia

72,355

Low Reference High

69,198 71,211 73,223

65,930 69,793 73,787

62,529 68,146 74,145

58,721 66,005 74,045

Lambton Shores

10,645

Low Reference High

9,919 10,502 11,085

9,157 10,260 11,434

8,313 9,862 11,611

7,392 9,307 11,595

Point Edward

2,030

Low Reference High

1,762 1,916 2,071

1,526 1,798 2,093

1,312 1,671 2,092

1,109 1,529 2,058

Rural Lambton

39,585

Low Reference High

37,799 38,643 39,487

35,905 37,521 39,178

33,854 36,165 38,593

31,556 34,469 37,597

St. Clair

14,515

Low Reference High

13,823 14,179 14,534

13,059 13,756 14,469

12,224 13,223 14,272

11,290 12,536 13,876

Petrolia

5,540

Low Reference High

5,469 5,792 6,115

5,399 6,034 6,713

5,299 6,248 7,324

5,142 6,410 7,928

Plympton-Wyoming

7,565

Low Reference High

7,192 7,448 7,705

6,824 7,308 7,811

6,416 7,115 7,863

5,959 6,846 7,830

Under 4,000 Rural

11,965

Low Reference High

10,913 11,275 11,638

9,900 10,563 11,250

8,925 9,842 10,823

7,965 9,085 10,323

Brooke-Alvinston

2,560

Low Reference High

2,294 2,443 2,592

2,042 2,316 2,609

1,801 2,178 2,606

1,570 2,028 2,581

Dawn-Euphemia

2045

Low Reference High

1,790 1,921 2,052

1,551 1,789 2,044

1,335 1,652 2,015

1,132 1,505 1,963

Municipality

2016 2021 2026 2031 Projection Projection Projection Projection

P a g e | 22

Population Projections 2006 to 2031- Lambton

Enniskillen

2,935

Low Reference High

2,538 2,716 2,895

2,179 2,499 2,839

1,869 2,299 2,788

1,592 2,099 2,714

Oil Springs

715

Low Reference High

611 681 752

509 636 778

428 592 796

359 549 808

3,710

Low Reference High

3,359 3,532 3,705

3,030 3,347 3,683

2,714 3,153 3,642

2,404 2,945 3,579

Warwick

Results Lambton County’s 2011 population is approximately equal to 2006. It did well in attracting ages 30-44 and did a better job of youth retention than expected. This raises projections slightly, however population decline in the long-term is still expected. Sarnia gained approximately 935 people and has shown some variability in growth rates in ages 50 and younger. It did well in attracting children and ages 40 to 49. It exceeded expectations in retaining or attracting ages 20 to 39. The influence of these recent census results raise long-term projections significantly compared to the 2006 projections and suggest the potential for modest growth, but decline remains most likely in the long-term. Lambton Shores has experienced a large amount of variability in growth rates unexpected decline following a census of strong growth. 2011 growth rates were well below expectations of 2006 projections in all age groups. The 2006 projections put much weight on 2006, which was a good year for Lambton Shores. The reversal in trends changes the long-term projections significantly – perhaps back to what would have been expected prior to 2006. 2006 projections were for Lambton Shores to maintain its population in the long-term with strong potential for growth. 2011 census results suggest potential to regain the 2006 population lost in 2011 but that long-term decline is more likely. Staff note discussion is included under the 2011 Census Results section regarding the possibility of there being some kind of quirk in the census data with respect to Lambton Shores. Point Edward experienced greater than expected growth in retirees and seniors but also ages 15 to 19 and 25 to 29. Its rates are quite variable and predict anything from significant decline to very modest growth. Rural Lambton has shown some variability in the attraction of age groups that would appear to be young families with children, but has otherwise had very consistent rates. Its long-term projections are essentially unchanged from 2006 – steady, but modest decline. Petrolia, Plympton-Wyoming, and St. Clair populations were all very close to what was predicted by 2006 projections. Their new projections increase very slightly. St. Clair’s growth rates have been very consistent, and indicate slight, but steady decline.

Population Projections 2006 to 2031- Lambton

P a g e | 23

Plympton-Wyoming is likely to decline, but shows potential to maintain or increase its population slightly with variability in age 35 to 39 growth rates. Petrolia rates continue to indicate growth, but show a significant amount of variation. As individuals, the rural municipalities with smaller populations show a high variability in growth rates due to their small populations. As a group (Under 4,000 Rural) however, they show consistent rates and a steady population decline that is expected to continue. All of these rural municipalities declined at rates on the lower side of expectations and it is felt the 2006 predicted rates were inflated because of the influence of being blended with the larger rural municipalities Plympton-Wyoming, St. Clair and Petrolia. In particular, this blending resulted in losses in young adults, and in some cases the middle-aged, being greatly under-predicted. Warwick and Enniskillen, in particular lost greater than expected population is ages 20 to 34. Varying degrees of population decline are expected for rural municipalities with small populations. Expected declines are even lower than 2006 projections. This is due first to the lower beginning population resulting from losses from 2006 to 2011 and second because of the switch to blending their rates with the Under 4,000 Rural entity, which eliminates some of the influences noted.

Future Age Distribution and Explaining Declines Despite relatively good results in the County as a whole and the City of Sarnia in particular, population decline in the long-term remains likely. Growth in particular age groups was quite good in comparison with previous years, and some municipalities are likely to see growth in the next few census periods, but the age distribution remains what will define future populations. Historic trends in age distribution were discussed under the section “2011 Census Results and Ongoing Trends”. Figures 3 and 4 showed how the baby boomer population has continued along from census to census at approximately the same size. Also, the number of young adults has remained relatively stable despite significant losses of children as they become adults. This has been sustained in the past by a large starting population of children. Figures 3 and 4, however, show that the number of children is significantly less now than in past censuses. Figures 12 and 13 show the future age distributions that can be anticipated for Lambton County and Sarnia if growth and decline in the various age groups continue at current rates. In the past, the baby-boomers’ age group aged with little change in over-all size. In coming censuses they will still make up a large proportion of the age distribution, but their number will begin to decline significantly due to their entering the ages of increasing mortality rates. The number of children can be expected to continue to decline, though not as great a decline as the last twenty years. This is tied to a decline in the child-bearing-aged population that can be expected over the next twenty years, though not at the same rate as when the tail-end of the baby boomers left child-bearing-years.

P a g e | 24

Population Projections 2006 to 2031- Lambton

Fig 12

Fig 13

The young adult populations stayed relatively steady in the past because of a large starting population of children that offset losses. With the decline in child-aged population over the last twenty years, a continuing high rate of loss of young adults would begin to result in noticeable declines in the actual number of young adults in the population. Discounting the inevitable and uncontrollable population losses resulting from the mortality of the baby-boomers, assuming 100% youth retention could be achieved, and taking into account immigration rates, these graphs suggest it is doubtful whether current birth rates are sufficient to maintain a steady population in the long-term.

Sensitivity Figures 12 and 13 clearly illustrate that the aging of the existing population and the current age distribution have a much greater influence on future population figures than the age-specific growth rates. An exception to this is the rates of loss in young adults. The 2006 projections included a sensitivity analysis that suggested future population projections would be most likely to increase if there are declines in the losses of young adults (youth retention), through the attraction of working age population and their children, and through increased birth rates. Conversely, the attraction of retirees would have less of a long-term impact on future population. The macro trends have not changed since 2006 and these conclusions are still applicable.

Monitoring These projections will have to be updated with each census. In particular, there will be a need to calculate the new age specific growth rates each census period and incorporate them into the projected future growth rates. The success of the projections in predicting growth trends and actual populations and the validity of the projections’ assumptions will have to be assessed with each census period.

Population Projections 2006 to 2031- Lambton

P a g e | 25

Adoption for Planning Purposes Table 1, summarizes the “low”, “reference”, and “high” growth scenario population projections for each municipality and each census period out to 2031. The “reference” scenario is our estimate of future population based on the assumption that future age-specific growth rates will be an average of the three most recent census periods. For rural municipalities with a population under 4,000, the age-specific growth rates are averaged with those of the geographic entity we have called Under 4,000 Rural. The “low” and “high” scenarios take into account the amount of historic variation that has been experienced in these age-specific growth rates, and the potential for projections to vary from the reference scenario.

Recommendation It is recommended that the projections as presented in Table 1 be adopted for planning purposes.

P a g e | 26

Population Projections 2006 to 2031- Lambton

Appendix A Projected 2031 Populations in Comparison with 2006 Projections In the graphs of Appendix A, green shows the populations that had been predicted for 2011 to 2031 by the 2006 projections based on 1991 to 2006 census data. Blue shows actual population and this report’s projections for 2016 to 2031 based on 1996 to 2011 census data.

Population Projections 2006 to 2031- Lambton Projected 2031 Populations in Comparison with 2006 Projections

P a g e | 27

P a g e | 28

Population Projections 2006 to 2031- Lambton

Appendix B Projected Age Specific Growth Rates Appendix B illustrates the average growth rates of the most recent three census periods (the “reference” scenario). The “high” and “low” scenarios illustrate the amount of variation in growth rates (the range or difference between the maximum and minimum rates) over the last three census periods.

Population Projections 2006 to 2031- Lambton Projected Age Specific Growth Rates

P a g e | 29

P a g e | 30

Population Projections 2006 to 2031- Lambton

Appendix C Actual 2011 Age-Specific Growth Rates in Comparison with 2006 Projections In the graphs of Appendix C, green shows the age-specific growth rates that had been predicted by the 2006 projections based on 1991 to 2006 census data. Blue shows the actual age-specific growth rates of the 2011 census.

Population Projections 2006 to 2031- Lambton

P a g e | 31

Actual 2011 Age-Specific Growth Rates in Comparison with 2006 Projections