The model illuminates the potential effectiveness of changes to planning policy in two ways. First, values for the Baseline scenario indicate the densities that might result under reasonable assumptions about future patterns of growth. Second, the seven alternative scenarios provide a sense of how much might be accomplished through reasonable alteration of different variables. Each is discussed in turn.
Densities produced in the Baseline scenario
Densities were calculated on four land bases: gross area, gross area net of natural heritage features, developable area, and net residential lot area. (See Fig. 52.)
The purpose of calculating density on the gross area net of natural heritage features is to simulate the rule specified in the Growth Plan (s. 2.2.7.3), which requires that the minimum density target be applied to the total designated greenfield area of upper- and single-tier municipalities, excluding features identified in provincial or municipal plans where urban development is prohibited. (See Fig. 1.) As natural heritage features are not evenly distributed across the region, municipalities with more undevelopable land would be penalized if the density target was applied to the gross land area. The Growth Plan policy levels the playing field. In the planning process, however, natural heritage features are often linked and buffers are added to create a natural heritage system. Local plans may prohibit urban development on these additional lands, yet they would still be part of the land base on which the minimum density target is applied. For this reason, densities are calculated for the gross area, gross area net of natural heritage features, and the developable area (i.e., the gross area net of the natural heritage system).
Fig. 52: Land bases
Fig. 53: Baseline scenario densities
| Level of natural heritage protection | |||
Density (per hectare) |
| Low | Medium | High |
Population | ||||
| Gross density | 49.7 | 39.0 | 34.1 |
| Gross density (net of natural heritage features) | 52.3 | 46.4 | 46.7 |
| Developable area density | 55.2 | 55.0 | 55.9 |
| Net residential area density | 115.2 | 115.2 | 115.2 |
Employment | ||||
| Gross density | 6.0 | 4.8 | 4.1 |
| Gross density (net of natural heritage features) | 6.4 | 5.7 | 5.6 |
| Developable area density | 6.7 | 6.7 | 6.7 |
| Net employment land density | 38.8 | 38.8 | 38.8 |
Population plus employment | ||||
| Gross density | 55.7 | 43.8 | 38.2 |
| Gross density (net of natural heritage features) | 58.6 | 52.1 | 52.3 |
| Developable area density | 61.9 | 61.7 | 62.6 |
Dwelling unit | ||||
| Gross density | 15.6 | 12.2 | 10.7 |
| Gross density (net of natural heritage features) | 16.4 | 14.5 | 14.6 |
| Developable area density | 17.3 | 17.2 | 17.5 |
| Net residential area density | 36.1 | 36.1 | 36.1 |
Bold values indicate combined population and employment densities per hectare greater than 50.
As might be expected, higher levels of natural heritage protection reduce gross density and gross density net of natural heritage features. (See Fig. 53.) While the Low case has a gross density of 55.7 residents and jobs combined per hectare, the gross density for the High case is 38.2. Considering gross density net of natural heritage features, the difference between the Low and High case is a drop from 58.6 to 52.3 residents and jobs combined per hectare, both greater than the Growth Plan target. The density of the developable area decreases slightly from 62.6 in the High case to 61.9 in the Low case.
Observations
First, the definition of the land base on which the minimum density target is applied affects performance. The more land that is "netted out" of the gross area, the higher the densities observed. If the degree of natural heritage protection is unevenly distributed across the metropolitan region, it may be easier for some municipalities to achieve the target than others. Testing this conjecture is beyond the scope of this project.
Second, the analysis suggests that while many urban areas developed in the postwar period fall short of the minimum density target, especially outside the City of Toronto (Mitra 2007:75-76; Mitra & Gordon 2007), the Growth Plan's target can be reached. As most of the density is delivered by the residential component, a fact consistent with the analysis in Section 2, achievement of the target largely hinges on increasing the density of housing.
Finally, as discussed in Sections 2.5, 2.6, and 2.7, it should be remembered that higher densities are a necessary, but not sufficient condition to bring about desired changes in transportation behaviour. The potential effectiveness of higher densities also depends on the degree to which the developable area is contiguous, as contiguous urban areas are more easily served by public transit and can be designed with more accessible street networks that facilitate walking and cycling.
Comparison of scenarios
A more interesting question than whether the scenarios produce densities above or below the Growth Plan's threshold is what magnitude of change is produced by altering different input assumptions -- in other words, what changes are likely to produce the most "bang for the buck"?
Changes to housing type mix
As Fig. 54 shows, shifting the housing type mix away from the single- and semi-detached categories towards townhouses, stacked townhouses, and apartments increases density. The number of dwellings on the net residential lot area increases by 13.3% in the Forecast Mix scenario, and by 20.3% in the Market Shift scenario. As the higher-density dwelling types are assigned smaller average household sizes, the net population density increase for each scenario is also smaller -- 10.4% and 15.9%, respectively. When considering the gross and developable area land bases, the impact of the change in housing type mix is further diluted. The Market Shift scenario, which increases net dwelling unit density by 20%, raises gross combined population and employment density by only half as much or less in each of the natural heritage protection cases.18
To explore this idea, a comparison with the post-1980 Section 2 study areas is in order. Why do these hypothetical cases have higher densities under the Baseline scenario than the 1980s-90s cases analyzed in Section 2? The residential lot area's share of the developable land area (46%) is comparable to the average observed in Section 2 (49%). Developable area employment density (6.7 jobs per hectare) is within the range of the 1980s-90s cases (2.9 to 8.1 jobs per hectare). It follows, then, that the higher densities produced by the model are delivered almost exclusively by the nature of the housing. In general, the study areas in Section 2 have a housing type mix more weighted towards single-detached dwellings than that used in the Baseline scenario (59%). With the exception of the Richmond Hill study area, which contains a high proportion of row and semi-detached houses (48%, versus 40% single-detached), the 1980s-90s study areas contain between 67% and 89% single-detached houses. The Baseline scenario's housing type mix assumptions therefore produce virtually double the net residential area density (115 people and 36 dwellings per hectare) than those observed in the 1980s-90s case studies (on average, 63 people and 19 dwellings per hectare). It may also be that the Baseline assumptions for lot size and average household size by unit type are more aggressive than those for the 1980s-90s study areas. Unfortunately, no comparison can be made using available data.
Fig. 54: The impact on density of changes to housing type mix
| Baseline | Forecast Mix | Market Shift |
Housing type mix | |||
Single-detached | 59% | 49% | 44% |
Semi-detached | 17% | 14% | 14% |
Rowhouses / townhouses | 17% | 21% | 21% |
Stacked townhouses | 2% | 3.5% | 4% |
Apartments | 5% | 13.5% | 17% |
Impact on population density |
| Forecast Mix | Market Shift |
Gross area & Developable area |
| Low + 3.5% Medium + 5.5% High + 8.8% | Low + 8.0% Medium + 10.1% High + 12.0% |
Net residential lot area |
| All + 10.4% | All + 15.9% |
Impact on population plus employment density | |||
Gross area & Developable area |
| Low + 3.3% Medium + 5.1% High + 8.1% | Low + 7.4% Medium + 9.3% High + 11.1% |
Impact on dwelling unit density | |||
Gross area & Developable area |
| Low + 6.3% Medium + 8.3% High + 11.7% | Low + 12.1% Medium + 14.2% High + 16.3% |
Net residential lot area |
| All + 13.3% | All + 20.3% |
Fig. 55: The impact on density of changes to public facility standards
| Baseline | Consolidated |
Standards |
|
|
Rights-of-way | 26% of developable land area | - 23% |
Parks and schools | The greater of: (a) Parks: 5% of land area + 2% of employment land area, or 1 hectare per 300 dwellings, whichever is greater; plus Schools: Public elementary: 2.5 hectares at 1 per 1,000 dwellings, Catholic elementary: 2.0 hectares at 1 per 2,600 dwellings, Public secondary: 6.5 hectares at 1 per 4,500 dwellings;
or
(b) Central Pickering value: | - 20% |
Impact on population and dwelling unit density |
| Consolidated |
Gross area & Developable area |
| Low + 16.5% Medium + 15.3% High + 13.2% |
Impact on population plus employment density | ||
Gross area & Developable area |
| Low + 15.2% Medium + 14.1% High + 12.2% |
Changes to public facility standards
In the Consolidated scenario, land allocation standards for schools and parks are reduced by 20% relative to the Baseline scenario and land coverage for rights-of-way is reduced from 26% to 20% (a decrease of 23%). As Fig. 55 shows, these changes increase densities relative to the Baseline scenario. Both population and dwelling unit densities increase by between 13.2% and 16.5%, depending on the degree of natural heritage protection, while combined population and employment density increases by between 12.2% and 15.2%.
Changes to natural heritage protection
Increasing the size of the natural heritage system affects gross density only. While the amount of developable land is smaller than it would otherwise be, its internal density remains constant. As might be expected, increasing the amount of undevelopable land decreases gross density. (See Fig. 56.)
Fig. 56: The impact on density of changes to natural heritage protection
Baseline | Green | |
Standards | ||
Natural heritage system | Designated natural heritage features plus buffers: Low 10% of gross land area Medium 29% of gross land area High 39% of gross land area | + 20% |
Impact on population, population plus employment, and dwelling unit density |
| Green |
Gross area |
| Low - 2.6% Medium - 10.0% High - 13.0% |
Changes to employment and mix of use
The final set of scenarios modifies the location of employment and the degree of mix of use. In the Mixed-Use scenario, the amount of employment land in each case is increased from 10% to 25% of the developable land area. At the same time, some jobs are moved out of employment lands into mixed-use settings, while the density of the remaining employment is increased by 25%. The combined effect of these shifts is to raise the jobs density on employment lands by 33%.
In the Jobs-Housing Balance scenario, the amount of employment land is increased until there is enough to support one job for each member of the employed labour force, assuming a participation rate of 60%. Keeping all other assumptions constant, this occurs when employment land occupies 36.8% of developable land, which, incidentally, is higher than that of any of the districts analyzed in Section 2.
As Fig. 57 shows, the result in both scenarios is a significant decline in gross and developable area population density. Also, as employment land is less dense than residential land in all scenarios, replacing residential with employment land reduces the overall density.
Looking at combined population and employment density, we see that, in the Mixed-Use scenario, the increase in employment density compensates for the decline in overall population density. This is not true of the Jobs-Housing Balance scenario, where the introduction of additional employment land reduces gross density by between 18.8% and 21.3%, depending on the level of natural heritage protection.
Fig. 57: The impact on density of changes to employment location and mix of use
| Baseline | Mixed-Use | Jobs-Housing Balance |
Assumptions | |||
Employment land | 10% of developable land area | 25% of developable land area | Optimizing for one job for every member of the employed labour force = 36.8% of developable land area |
Jobs density on employment lands (per hectare) |
Business/Industrial Parks: 40 Major Office: 100 Retail: 50 | + 25%: Business/Industrial Parks: 50 Major Office: 125 Retail 62.5 | -- |
Job mix | Mixed-use settings: 18% Business/Industrial Parks: 50% Major Office: 20% Retail: 12% | Mixed-use settings: 28% Business/Industrial Parks: 41% Major Office: 25% Retail: 6% | -- |
Impact on population density | Mixed-Use | Jobs-Housing Balance | |
Gross area & Developable area | Low - 24.8% Medium - 22.7% High - 26.2% | Low - 42.6% Medium - 43.6% High - 44.9% | |
Impact on employment density | |||
Gross area & |
| Low + 188.9% Medium + 189.8% High + 187.7% | Low + 176.2% Medium + 176.2% High + 174.7% |
Net employment land area |
| All + 33.0% | -- |
Impact on population plus employment density | |||
Gross area & |
| Low - 2.6% Medium + 0.4% High - 3.1% | Low - 18.8% Medium - 19.7% High - 21.3% |
The Big Moves scenario
The Big Moves scenario, which combines the Market Shift, Consolidated, and Mixed-Use scenarios, produces an increase in density on all land bases. (See Fig. 58.)
Fig. 58: The impact on density of the Big Moves scenario
Impact on population density | |
Gross area & | Low + 30.4% Medium + 29.7% High + 25.7% |
Net residential lot area | All + 15.9% |
Impact on employment density | |
Gross area & | Low + 45.2% Medium + 45.0% High + 43.8% |
Net employment land area | All + 33.0% |
Impact on population plus employment density | |
Gross area & | Low + 32.0% Medium + 31.4% High + 27.6% |
Impact on dwelling unit density | |
Gross area & | Low + 35.4% Medium + 34.7% High + 30.5% |
Net residential lot area | All + 20.3% |
Observations and implications for policy
Comparing the scenarios leads to the following general observations and conclusions:
First, the different scenarios produced a wide range of outcomes. For example, total population ranged from 7,500 in the Jobs-Housing Balance / High natural heritage protection scenario to 25,900 in the Big Moves / Low scenario. The Baseline / High scenario produced 1,650 jobs while the Mixed-Use / Low scenario resulted in 7,000. Gross population-plus-employment density ranged from 30 residents and jobs combined per hectare in the Jobs-Housing Balance / High scenario to 74 in the Big Moves / Low scenario. In conjunction with the study area characteristics observed in Section 2, these ranges illustrates the wide range of outcomes that are possible on any given piece of land.
Second, both shifting the housing type mix towards higher-density dwellings and reducing public facilities standards can help increase overall density, although the latter may have a somewhat larger impact. Reducing public facilities standards by about 20% raised population-plus-employment density by 12% to 15%. Increasing net residential dwelling unit density by 20% raised population-plus-employment density by only 7% to 11%. The Big Moves scenario shows that these changes in combination can produce a substantial increase in density.
Third, in the absence of an offsetting increase in residential density, greater intermixture of residential and non-residential uses reduces density at the local-area scale. As the density of employment (jobs density) is typically less than population density, decreasing the land area of residential areas in favour of more employment land reduces gross and developable area density. The Mixed-Use scenario produced a marginal increase in population-plus-employment density only by redistributing the location of jobs and increasing their density. The Jobs-Housing Balance scenario, which increased the proportion of the developable area occupied by employment lands from 10% to 37%, resulted in a population-plus-employment density reduction of approximately 20%.
It should be recognized, of course, that the Growth Plan target is intended to be applied over the entire envelope of land designated by municipalities for future urbanization, not smaller-scale areas. A high-density node in one location may offset a low-density business park in another. But the Mixed-Use and Jobs-Housing Balance scenarios illustrate, however, that the creation of more mixed and more "complete" communities at the secondary plan scale may reduce local-area densities to below levels generally believed to support high-frequency public transit.
Other aspects of urban form should not be ignored. Increased natural heritage protection reduces gross density. As noted in Section 2.3, the exclusion of land from urban development by conservation authorities and other public agencies must be balanced against the need to create contiguous urban form and street systems that support the effective provision of transit, walkability, and other objectives.
Notes
18. Gross and developable area densities vary by degree of natural heritage protection because of the formula for quantifying public facilities land. The formula used in the model allocates parkland in proportion to a combination of dwellings and employment land, and different classes of schools, each consuming different amounts of land, in proportion to dwellings. The result is a non-linear relationship. Public facilities land per thousand people therefore ranges from 2.21 hectares in the Low/Consolidated scenario to 3.42 hectares in the High/Jobs-Housing Balance scenario. Were they quantified in direct proportion to population, there would be no difference. See Appendix C.2 and Fig. C.22 for more details.