How much might the CPC childcare credit cost and where is the rest of the federal ELCC money going to go?
By Jennifer Robson, Associate Professor of Political Management, Carleton University. My disclosure statement is posted on my university webpage here.
*This story was updated August 21 after I received a table produced by the Conservative Party of Canada that I have not previously seen.
A few days ago, a few different journalists called me and asked me some smart questions about the CPC platform, including the proposal to “convert the Child Care Expenses Deduction into a refundable tax credit”.
In Fall 2014, I warned that leaving the Child Care Expenses Deduction (CCED) unchanged (plus the state of child benefits at the time) might undo a lot of the good that the NDP’s childcare proposal could otherwise do. This is still a potential problem, IMHO…. Ahem….
I’m also on the record here and here arguing that the CCED should be turned into a progressive refundable tax credit (RTC). Advocates, please take note that the recommendation in the second piece was to *combine* significant supply-side spending with a progressive RTC instead of the CCED. As Prof. Rob Gillezeau noted, you can be in favour of BOTH CCED reform AND strong supply-side public funding. I have never argued that an RTC will create spaces. It won’t. But it could help with the out of pocket costs for families. Even if we get fees down to an average of $10/day, paying that will still be difficult for many families. Arguments for keeping the CCED as-is too often come from very privileged vantage points.
Now that Mr. O’Toole has confirmed his intention to cancel the new bilateral ELCC agreements signed with 8 subnational governments (including Nova Scotia where the recently-elected Progressive Conservatives have said their government wants to keep that deal), journalists asked me if a new RTC would really cost as much as those agreements. That’s what this piece is about.
Remember those new agreements are worth $29.8 billion over the next 5 years and $8.3 billion ongoing.
The CPC platform hasn’t yet been costed so I look forward to and will trust the work of the Parliamentary Budget Officer (PBO) when it comes out.
In the meantime, I’ve been trying to work out if it’s possible to repurpose CCED spending, plus nearly $30 billion in ELCC funding over 5 years and at least $8.3 billion ongoing, all into the type of credit the CPC platform outlined. Or is squeezing all this money into a new tax credit as difficult as passing a camel or an elephant through the eye of a needle?
Let’s walk through this together, step by step. This is going to be a bit rough, but it’s a start while we wait for more details from the party and PBO.
Step 1: Set your baseline.
The CCED is a deduction that eligible taxfilers can claim for eligible expenses in the previous tax year. You can read the myriad rules on eligibility here. The deduction reduces an eligible claimant’s taxable income before federal and provincial taxes (outside QC) are calculated. The cost (in foregone federal tax revenues) of the deduction goes up and down a bit, especially in a pandemic year when providers closed and families stayed home, but it is generally about $1.4 billion a year (note: tax year, and no I’m not futzing about here converting that into fiscal year). Here’s what the trend looked like in the 2020 and 2021 federal Tax Expenditures and Evaluation Reports:
The most recent of those reports has a very good and lengthy evaluation of the CCED that I strongly encourage you to read. (Disclosure: I have been a peer reviewer on other sections of the annual report for the last few years. I was paid a small stipend in one year and declined it going forward). I’ll be drawing on numbers from it later, but here are some jaw-droppers:
- While 6.4 million tax-filers (you have to file a tax return to get the CCED) have a child, just 1.4 million are eligible for and claim this deduction.
- Tax-filers in families with incomes over $150,000 per year make up 13.4% of all CCED claimants but receive 51.3% of the $1.4 billion in tax savings.
Step 2: Describe the policy change
The CPC proposal is to convert the CCED into a refundable tax credit. Refundable means, like the GST/HST credit and the Canada Child Benefit, it’s payable as a cash benefit, separate from what you may owe or save in income taxes. The CPC platform included a few illustrative examples of the projected value of the new credit and we can use those to estimate what the benefit rate would be. And on August 21, I saw a copy of a table that the party had circulated to a few people. Here’s what we now know:
- It’s using the CCED maximum eligible expenses amount ($8,000 per year for a child under 7, amounts are lower for older children and higher for children with a disability).
- The maximum value is 75% of $8,000 (or $6,000) and it declines with income, reaching a constant of 26% when family income is $162,975 or more. I’m going to round this to $163,000 for the sake of simplicity.
- Like the CCED, it relies on what families report paying in practice (subject to the upper limit). If a family eligible for the 75% rate reports $4,000 in eligible ELCC expenses, they receive 75% of the $4,000, not 75% of $8,000.
- It is using family income, rather than personal income. The CCED uses personal income but says in a couple the *lower* income parent has to claim the deduction (note: deductions are worth more to higher income tax-filers because of our progressive rate structure).
Here’s a graph showing the benefit structure based on the data table I was able to see on August 21. Note: I’m using categories of $10,000 in family income here plus an extra category for those with $163,000 or more when the proposed CPC RTC takes on a constant value, as far as I can tell, regardless of family income.
Step 3: Be clear on your assumptions.
Here are the assumptions I’m making in this rough costing effort:
- The CCED definitions of eligible expense ELCC types stay, along with annual maximum amounts. So too does the CCED rule saying expenses can be pooled among kids in the same family to maximize annual amounts.
- Like other RTCs administered by the Canada Revenue Agency, family income is Adjusted Net Family Income (ANFI) rather than gross, taxable or net. Because ANFI deducts CCED amounts from income and that would create a circularity problem here, I’m using an amended definition to deduct pension contributions and union dues but not ELCC expenses.
- The credit, like other RTCs, is not taxable and is paid after the value of any provincial subsidies (ie: it is on actual out of pocket costs, not the sticker price). I also assume these costs are accurately reflected in my microdata source.
- Administrative obstacles to collecting the CCED (the need to file a return, the need to know about the tax measure, the need to retain receipts, the rules on reporting the SIN of an individual provider, the rules on types of care, rules on which parent in a couple can claim it, etc..) are unchanged.
- There is no behavioral response from tax-payers. Note: this is a contestable assumption. Maybe making the deduction into a credit spurs more people to claim it, driving up the take-up rate above the current CCED. Maybe. But if so, I bet not in the short-run. If your type of childcare isn’t CCED-eligible, do you really change providers for a tax credit? Can you even find another provider? If you don’t claim the CCED because you don’t think it’s for you, or can’t get receipts from your provider, does making this a refundable credit change that?
- I’m also assuming that, like the CCED, there is a per-child amount, but the total claim cannot exceed 2/3 of income. For the CCED that’s the personal income. Here I assume that threshold is 2/3 of family ANFI.
Step 4: Describe the target population
I don’t have the micro-simulation tools used by the PBO, so instead I start by looking at the Canadian Income Survey (Public Use Microdata File) for the most recent year available through my university (2017). It has the relevant income variables, generally linked to actual tax returns, plus basic demographic and self-reported out-of-pocket costs for ELCC. I get a total of 5.5 million families with a child and self-reported ELCC costs of $1 or more. I work out the median ELCC expenses for families, again grouping families by income (ANFI-like definition) category from up to $10,000 through to $163,000 or more. Note, this includes a mix of couple and lone parent families, and families with different numbers and ages of kids. My estimates reflect all of those differences.
Here, for example, are the median ELCC costs reported by couple families for a young child.*
Then, I work out the *potential* value of an RTC for all of these families, using the estimated parameters of the CPC policy proposal, *if* all 5.5 million families with ELCC expenses claim the benefit. I get a value of $19.7 billion per year in 2017 dollars, or about $21.2 billion in 2021 dollars. BUT, the CCED upper limits are NOT inflation adjusted so any change over this time would be due to changes in ELCC costs, family behavior, and demographic changes.
Anyway… let’s just go with $21.2 billion as our estimate for 2021 if EVERY family got up to 75% (actual amount depends on income) of their actual childcare expenses (subject to upper limits) in a new RTC.
Step 5: Adjust your estimate for expected take-up
We know that the need to file a tax return, get and keep receipts, know to claim a benefit, find and use eligible childcare, etc…. all reduce take-up of the CCED. We can’t observe those factors directly, so we have to estimate. As we saw above, Finance says not all families with ELCC costs claim the CCED or are even eligible for it. So let’s reduce our 5.5 million CIS families paying ELCC fees down to something that looks more like the take-up rate of the CCED — that’s 1.4 million.
Here I now assume that those 1.4 million CCED users are evenly distributed in the data underlying our initial estimate of $21.2 billion. This is a strong assumption. Lower income families are less likely to file a tax return, more likely to report lower ELCC expenses (see both Finance’s evaluation and the CIS data above), and judging from the Finance evaluation, may be less likely to qualify for the other administrative rules associated with the CCED. For an RTC with a somewhat progressive rate structure, this could all mean that what you ‘save’ in having lower benefit rates for middle and higher income families, you ‘loose’ in paying those rates on bigger individual claims.
So, if take-up is likely to be something more like 25% of the potential population, then our estimate of costs drops from $21.2 billion to $5.94 billion in 2021.
Step 6: Look at costs, over time, relative to earmarked fiscal room
Now we pull this all together. I’m going to start in 2021 and adjust for inflation at 3% next year and 2% thereafter (yep, I think current inflation rates are probably transitional), but I’ll use the ELCC transfer amounts booked in Budget 2021 (since these are nominal) and the 2021 and 2022 projected values for the CCED from Finance’s Tax Expenditure Report.
As far as I can tell, the CPC policy would pull $8.7 billion out of the ELCC system over the next 5 years. This would largely come out of provincial coffers, resulting either in a substantial reduction to publicly-announced plans by sub-national governments of all stripes, or big new fiscal holes if those governments try to keep their promised ELCC changes.
**Because I did this same exercise previously excluding families with incomes over $150,000 and came up with a figure of a little more than $16 billion, we can conclude that about half of the reduced ELCC spending (relative to the current federal benchmark fiscal plan) would go to families with incomes of $150,000 or more. This is replicating, NOT reducing, the distributional problems of the CCED. **
From reading the CPC platform, I don’t know how that $8.7B money would be used instead. But it doesn’t look like it will be on ELCC. And from what I can see, in repurposed spending, sure looks like it flows to parents who can already afford to spend the most on childcare.
Addendum:
It is possible to walk and chew gum at the same time. It is possible to reform the CCED AND keep agreements on intergovernmental transfers.
It is also possible to design an RTC in a number of different ways. In the table below, I provide some brief information and links to learn more about how the three existing income-tested ELCC credits work in each Québec, BC and Ontario.
Online calculator for Québec credit: http://www.budget.finances.gouv.qc.ca/budget/outils/garde-net-en.asp
Online calculator for BC credit: https://myfamilyservices.gov.bc.ca/s/estimator
Online information for Ontario credit: https://www.ontario.ca/page/ontario-child-care-tax-credit#_How_the_tax
Acronyms:
CPC : Conservative Party of Canada
ELCC: Early Learning and Child Care
CCED: Child Care Expenses Deduction
RTC: Refundable Tax Credit
CIS: Canadian Income Survey
ANFI: Adjusted Net Family Income
IMHO: In my humble opinion.
*The CIS PUMF reports the age of the youngest person in the family as either 0–5, 6–9, or 10–15. The CCED uses age ranges of 0–6 and 7–15. I limit families to those with children younger than 16. The impact of the difference between the CIS and CCED categories for younger children are negligible given how few families report expenses at or above the maximum threshold.
