3.3. Charting Revenue Results, 1990-2020
In order to rebuild better, it’s important for Council and the public to see how we got here; the purpose of this article is to anchor municipalities’ annual fiscal squeeze and occasional crisis in their historic context.
For the last 30 years, three sources dominate Ontario municipalities’ revenue actuals:
1. Property Tax, (including real property tax, business property tax, special assessments and payments in lieu of taxation);
2. Fees & Charges, (all sales of goods and services); and
3. Transfers, (all sources and types).
Statistics Canada’s is the most authoritative warehouse for government finance data by sector. The following slides show extracts of its municipal revenue data for Ontario in relative and actual terms, showing how the composition of municipal revenue changed over time in terms of the shares from each source, nominally and year over year.
The following charts provide an overview of 30 year trends in the types and top sources of Ontario municipalities’ revenue.
Takeaways
T R A N S F E R S
Volatility
The one-year change in nominal transfer revenue as a per cent of total transfer revenue was in the double digits 12 times in 29 years. Nominal year-over-year fluctuation on this order of magnitude happened once in the two other main sources of revenue - tax in 1998 and sales of goods and services in 1999 (Chart 1 below). Per slide 5 in the overview, transfers as a share of tax ranged 56%, though this range tightened in in the latter half of the study period.
Chart 2 illustrates how transfer peaks and valleys ripple across the revenue profile and are mirrored in the shape and variability of total revenue - the largest contributor to year-to-year fluctuation.
2. Influences on supply of intergovernmental transfers
The timing of provincial and federal elections and the ideological response of each sphere to its economic and financial circumstances (based on a 5-6 year cycle) appear to be the dominant influences on grant activity. Federal transfers to provinces also influence provincial transfers to municipalities. (Echoing YIELD 3.2.1 on data structure, Statistics Canada last disaggregated transfers to municipalities by operating and capital, federal and provincial, conditional and unconditional and service area in 2008, so no charts are shown.)
We can still extrapolate that, based on these pre-2008 influences, there is a risk that the impact of COVID on municipal revenue is double-barreled. Initially, weak demand for services resulted in unrealized user-driven revenues. Over time, as more federal and provincial resources are directed to servicing pandemic debt, softening transfers to municipalities is one scenario to plan for.
2. S O U R C E I N T E R A C T I O N S
Absent local policy direction, municipal revenue decisions can be a reaction to external spending decisions.
Transfers, expenses for the governments supplying them, are generally announced before municipalities' adopt a budget to give Councils a choice about how much of the transfer variance they want to absorb through expense management or local revenue supplements. High fixed costs limit municipal capacity to do less when they get less though, so, unless planned for or otherwise interrupted, the top job for own-source revenues can be shock absorption.
Chart 3 shows how, as shares of total revenue, transfers ranged from 37.1% and 20.4% over ten years, in 1992 and 2001, yet the combined total of tax and transfers produced a range of less than 10% of total revenue over three decades, worth between 66.6% (2019) and 75.3% (1993) over that time.
Charts 4 and 5 highlight how transfer changes factor into total revenues and effect own sources.
Firstly, there appears to be a two-year lag in how quickly own source increases compensate for transfer decreases: Transfers decreased in 1996 and 1997, and after an increase in 1998, they were back down below 1997 levels in 1999 and 2000. Own sources are on the same relatively flat incline in 1996 and 1997 as the preceding years; 1998 is the year they reflect the transfer change.
The delay suggests a reason why the boom and bust of the 1996-2000 period is unparalleled in total revenue results elsewhere in the timeline: Outside Premier Harris’ tenure, transfers grew at a slower pace than own source revenues, to be sure, but at no other point were there two consecutive years of nominal decreases.
Secondly, in terms of how own sources level off transfer volatility, it appears that tax shoulders most of the impact in year one, with the revenue impact spreading out across the other two own sources in that year and more so the next year.
1998 was the biggest one-year change in own source revenue in 29 years; 82.8% of the increase that year was from the property tax base. In the short term, to the extent that shifting federal and provincial priorities result in shifting funding to local government, the financial impact is disproportionately borne by property taxpayers. In other words, municipal property taxpayers pay for policy shifts elsewhere.
1998 shows a 37.7% one-year increase in property tax revenue and increasing proportions of revenue from the other two own sources that year - sales increasing 10% and other 12.3% from 1997 to 1998 - but more so in 1999, where sales increased by 18.2% year over year and other own sources increased by 14.6%. This is one possible reason transfer decreases in 1999 and 2000 don’t produce substantial movement in tax: by then, the other sources have caught up. Combined, they accounted for 26.2% of total revenue in 1995, 27.8% in 1996, 29.9% in 1997, 27% in 1998, 30.8% in 1999, 32% in 2000, 33% in 2001 and 33.3% in 2002.
As a share of total revenue, 1998 isn’t the peak year for tax, 1999 is, representing 47.3% compared to 45.7% in 1998. It declines slightly over the next three years to 45.1% in 2002.
Broadly, the fiscal behaviour modelled indicates that tax moderates sustained transfer losses and other own sources moderate tax over a multi-year period.
As cushions for grant volatility, yields from individual own sources rarely decrease from one year to the next: once it happened in tax, twice in sale of goods and services revenue and eight times in other own sources, (which, as discussed below, reflects the interaction between local rates and market demand and whose yields are the least Council-controlled). Rolled up, own source revenues only decreased in two years: 1996 and 2000.
2. Learning from our history?
It is difficult to reconcile a history of transfer volatility with the popularity of municipal lobby groups’ requests to increase funding to local government.
On the revenue side, nominal transfers in 2000 were lower than they were at any time from 1990-1996. The track record of fluctuating intergovernmental funding challenges the idea that, outside of emergencies, a larger external presence can play a stabilizing role in local budgets.
As long as the flow of tasks and requirements from the federal and provincial governments to municipalities is unencumbered, balanced budget mechanics at the municipal level mean that they give rise to a non-discretionary own source revenue requirement on the same order.
Therefore, the external presence in municipal operations drives own source revenue needs as compensation for transfer volatility in a high fixed cost environment and as a source of ongoing and incremental spending requirements to offset. Council interest in balancing out the external presence with a transfer equalizer is understandable; it is also ahistoric to expect the fundamental source of uncertainty, financial and otherwise, to be a source of consistency.
To mediate the effect of transfer change on the local tax base, municipalities receiving operating funding and those with grant-maximizing Councils may choose to link external contribution levels to proportionate tax rate stabilization reserve transfers.
3. A substitution effect?
It is possible that high transfers in the early 1990s offset the need to cultivate revenues from own sources, with the foregone revenue compounding the effect of declining transfers on own source needs a few years later. Chart 6 illustrates this per source, below.
Chart 4, above shows its effect on aggregate: the mustard waterfall chart in the middle shows very similar slivers of change from year to year for the first seven years.
4. Property taxes increased almost three times as fast after 1998 then they did before it
One effect of compounding financial pressures on municipalities during Premier Harris’ tenure (decreasing transfers, downloading services and changing property tax and assessment policy) is rapid appreciation in annual tax revenues after 1997. From 1990-1997, the median growth rate for nominal tax revenue was 1.6%. For 1998-2020, it was 4.3%.
After 2002, the service environment and transfer levels were relatively stable, so one interpretation for why tax revenue escalated so significantly from then on is the effect of provincial retention of policy control over municipal expense drivers, with changes then reflected in own source needs.
This is significant because Harris’ time in office is understood to be the most dramatic period in municipal operations and finance, however Chart 6 shows how his legacy also might be manifesting as a significantly higher growth rate in own sources for the latter 20 years of the study period.
5. Diversification first by force, then by choice
Diversification of own-source revenues usually refers to the pursuit of non-tax revenue. Change in revenue from sales of goods & services and other own-source revenues was minimal over 30 years, ranging from 17% (1992) to 23.2% (2002) and 7.3% (1993) to 11.9% (2019) respectively as shares of total revenue.
Looking at both non-tax sources, from 2011-2019, the combined average as a share of total revenue was 33.3%; this exact average was recorded for the years 2001 and 2002. Initially, given the speed of the increase, it is plausible that non-tax revenue hit one-third of total revenue on account of a lack of alternative funding or revenue sources in the face of growing expense pressure in the late 1990s. (At that time, transfers were declining and own-source revenue-raising powers were substantially curtailed in 1998.) In other words, diversification looks to have been pursued by necessity, not by choice. In a sense, financial dependence on provinces through grants was mediated by tax in the immediate term, then traded for higher reliance on the service market for consumption-based revenue. Ten years later, with several consecutive years with a combined average of above 33% of total revenue, higher sustained use of this group of lower-yielding sources appears to be a local policy choice. The changing composition of total revenue from 2019 to 2020 illustrates the vulnerabilities in higher reliance on demand-driven revenues - notably other own sources.
This forecasts the continuation of property tax as the mainstay of municipal revenue given that transfers and other own sources have a history of variability, with, per chart 3 in the overview, transfers showing the most dramatic fluctuations, other own sources the second most variability, most of which is concealed until the year-over-year view, and sale of goods and services more variable than tax but not by much in terms of total revenue, more in terms of change from year to year. Chart 7 bears this out, with property tax increasing 4.2 times its 1990 levels by 2019.
6. Market, demand and user-driven revenues are more variable than tax, exposing a higher share of revenue to fluctuation risk
Throughout this period, municipalities were encouraged to move whatever services they could off of the tax base onto user fees and charges and adopt full cost recovery pricing where possible. This was supposed to rationalize service costs and reduce municipal reliance on a sole source of locally-controlled revenue - tax. In 1990, the share of total revenue occupied by non-tax, non-fee own sources was 10.8%, so ending up at 11.9% before the pandemic is a negligible change. The nominal dollar view sheds light on why that may be: It was the only category whose growth rate shrank from 2019-2020, by 7%. Development charges are the biggest single source within the ‘non-tax, non-fee’ composite category, with substantial year-over-year fluctuation in nominal receipts corresponding to the 24-month building development cycle. Thus some measure of diversification is co-created between policy-makers and the market, and not always possible.
Without a multi-year financial planning capacity in which to set rates that build in a reserve component to pay for down years, it may not be desirable, especially for municipalities that set rates and adopt budgets one year at a time. Even in big cities that leverage their mass for revenue, with demand subject to fluctuations from a wide range of factors and high fixed costs, cashflow challenges, like those experienced in 2020, were not unforeseeable. The question is whether aggressive multi-year planning could have negated 2020 losses. In Toronto, this might have been possible; in places like Mississauga, it might not have been.
In cases where municipalities diversified out of necessity, the sequence of events in the intervening years starts to map the cycle of dependence:
Transfers were down from 1996-2020; tax room was saturated in 1998, so revenue needs were spread out amongst more sources; the sources available were more wobbly than tax revenue; one year, the user base dried up and the user fee revenue with; after depleting their own reserves, municipalities lobbied other governments for emergency relief; the funding that materialized arrived late, it was distributed in part to municipalities that didn’t need it, and insufficient for those that did; following the events of 2020-2022, no fundamental changes in the provincially-determined parts of the municipal revenue system are planned.
Where this cycle is a problem, taking local ownership of a plan to mediate it is a natural next step.
Methodology
1990-2020 was a dynamic period in municipal history: The number of years in a municipal term of office, number of municipal governments in the province and types of them, the legislative basis for municipal authority, expense mandate and financial reporting standards all changed in this time period.
The administrative units included in 1990-2008 data are municipalities, local non-profit housing corporations, improvement districts, and conservation authorities. The units in the 2009-2020 data are municipalities, conservation authorities, social service administration boards and various special-purpose bodies.
Revenue data recorded according to pre-2009 accounting were reconciled with data reflected according to post-2009 rules through two choices:
There are two strings of data for 2008; the pre-2009 version was used.
The two categories of revenue data for each period were reconciled as follows:
1990-2008
Property tax: Real property taxes, special assessments, business taxes, grants in lieu of taxes, other property-related taxes and land transfer taxes.
Sales of Goods and Services: Water, rentals, concessions and franchises and other sales of goods and services.
Other own source revenue: Lot levies, other licences and permits, remitted trading profits, interest income from own enterprises, other investment income, other fines and penalties and miscellaneous revenue from own sources.
Transfers: General and specific purpose.
2009-2020
Property tax: Real property taxes, special assessments, business taxes, payments in lieu of taxes and other taxes on property.
Sale of Goods and Services: Community amenities, rentals and parking, other sales of goods and services.
Other own source revenue: Taxes on use of goods and on permission to use goods or perform activities, property income, fines, penalties and forfeits, voluntary transfers other than grants and miscellaneous revenue.
Transfers: Grants from general government units.
Because various reconciliations and data refinements were needed to produce 30 year nominal data, we resisted a further intervention to inflation-adjust the data to produce real values.
Here is the worksheet where the streams of data are available in raw, reconciled and analysed forms.
Some 2020 data were preliminary and/or revised at the time of writing, and are prone to ongoing revision.
These factors emphasize that revenue actuals are subject to an unlimited number of possible influences and, as pieces of a big, moving whole, interpretations should not be considered conclusive or causal, especially as it pertains to changes within individual municipalities, which are concealed in sector-wide data.
Series data are estimates of aggregate results. According to Statistics Canada, preliminary data are collected from a representative sample of Canadian municipalities, 13% of all municipalities, and then “most” are cross-referenced with audited financial statements.