I’ve been following the reaction to the recent announcement by Liberal leader Justin Trudeau on proposed changes to federal personal income taxes and child benefits. Lots of very smart people have already weighed in both to critique (see Stephen Gordon here, Jack Mintz here and Andrew Coyne here) or to offer support (see Ken Battle and Sherri Torjman here) or cautious optimism (see Lindsay Tedds here). Both Mike Moffatt and Kevin Milligan have written supportive pieces, informed in part by the technical advice they gave Trudeau’s team. You should also read Tammy Schirle here on the costs of universal entitlements.

In the interests of disclosure, yes I have given some technical advice to folks working on the LPC platform and saw an advance copy of what they released. Yes, I was a federal Liberal staffer a very long time ago. I’ve also been a public servant under a Conservative government. I’ve spoken at a community NDP event. Once, I peeled vegetables for (now) Green Party leader Elizabeth May. I don’t believe that anyone with partisan experience is incapable of independent thought.

I’d like to take a closer look at how the 3 main federal parties are each carving out different policy turf when it comes to helping different families with kids. It’s not that I don’t care about costs, or growth, but I think it’s worth looking at household-level impacts. More specifically, I’m interested in what happens for kids under the different options on offer.

All 3 parties are staking claims that their policy package will do the most for ”œmiddle-class” families with kids. Below, I run the policy plans through different representative families to see who ends up with what.

The ”œmiddle-class” (whatever that really means) is a pretty inclusive  (maybe too inclusive) category.  Not all families in it look the same or have the same resources. The go-to proxy for membership in the ”œmiddle-class” (household income) is really, really different if you’re talking about 1 versus 2 earners in the household.  Finally, our three parties are offering different benefits depending on the number and ages of kids, use of daycare and household earners.

So I’m going to create several ”œtypical” families. I’m going to look at families with different numbers of kids and with both 1 and 2 parents working. I’m going to give them some income and daycare costs based on what’s average for families like them. Then we’ll see how they do under each of the three options on offer.

I’m not going to make any assumptions about how these families might change their childcare, how much they work or who they live with. Of course those things might change as families react to the programs and services that governments offer – though not all families have the same flexibility. But lets just deal with what families start with and what they’d get from each of the party platforms. Afterwards we can start to talk about behavioral responses.

First, we have to make some families

There are about 6.2 million kids aged 0-17 in Canada living in 3.4 million families.[1] Who those kids live with shapes how much money their family might have and, in turn, what each of the parties is offering. I’m going to create 8 prototype families to illustrate these differences.

Households grow (and shrink) and change over time, so the best we can do is take a snapshot based on Statistics Canada data. Broadly speaking, more kids live in a family with 2 parents who both work than any other family type. About 20% of kids live with a single parent and about a quarter of kids live in a 2-parent but 1-earner family. Family size is usually smaller with one-earner families and single parents are more likely to have kids older than 6. So, let’s create 8 families with kids that are prototypes for those different arrangements.

So now we have 8 families with kids and we know something about roughly what share of kids live in families like each of these:

Our 8 "typical" family types and the % of kids who live in them
Our 8 “typical” family types and % of kids who live in them, click to enlarge

And now let’s give our families the median earned income, based on their family type.[2] For dual-earner families, let’s split their incomes 60/40 which is the best way I can think of to account for both continued gendered wage gaps and the increased matching of couples by education level. I’m going to use earned income because it’s the best way to figure out a starting point BEFORE our families get various transfers under each plan. It’s also a better measure of the resources the family brings from the market when they look at different policy options offered by political parties.

Median earned incomes by family type
Median earned incomes by family type, click to enlarge

Does this make them each ”œmiddle class”? I have no idea. But it makes them “middle income” for their family and gives us a set of numbers to work with. I feel better about using these than the model families chosen by each of the parties for their marketing materials.

And finally, let’s give each of our families, with their earned income, an annual out-of-pocket daycare cost, based on national survey data.[3]

IRPP_Figure 3
Out-of-pocket daycare costs, based on mode, click to enlarge

Next, the options on offer:

None of the parties has yet released their full platforms. But they have all put a few wares in their store windows and told Canadian families they will be better off with their brand of policy compared to the other options. Here’s a quick summary to refresh your memory:

Summary of party policy (so far)
Summary of parties’ policy (so far), click to enlarge

Recognizing there’s still more to come, how would each of our composite families do under each party? If the goal is more dollar value for kids in families, how much could each of our families expect?

All costs are based on annual amounts and are for 2015. Let’s put all of our families in Ontario because it’s the most populous province. Let’s also assume that the NDP’s $15/day program won’t ask parents with a subsidy to pay more than they are now. Let’s assume that our families take full advantage of each program offered (which may not be true). Finally, I’m going to model the LPC plan with the Child Care Expenses Deduction (CCED). They haven’t said anything yet about what they would or wouldn’t do with it but they made some positive noises when the limits were increased. If I just take the CCED out of play, then I won’t create an accurate picture of what’s on offer.

With me so far? Eight ultra-representative families at their respective median earned incomes, with their respective average daycare costs and 3 different policy mixes, (under favorable assumptions), that each promise to make them better off…. Let’s see what happens!

IRPP_CPC
click to enlarge
IRPP_NDP
click to enlarge
IRPP_LPC
click to enlarge

And now, as promised, a graph comparing the top-line results across the parties, only this time, I’m going to put the family size, number of earners and family income back in to make reading a little easier:

IRPP_topline
click to enlarge

Ok, so there’s a lot to puzzle through in that chart.

Each party is promising Canadian families with kids that their package is best for their children, so let’s look at the options from the vantage point of those kids.  I’m going to suggest that, for kids, ”œfairness” will mean policy doesn’t judge them for the number of parents and earners they live with. Likewise, for kids, ”œefficiency” will mean getting more value of out taxpayer dollars by directing them where they are really needed and will make a real difference (ie: families that don’t have as much earned income).

The NDP package looks attractive if you’re a kid in a household  paying high daycare fees (families C, D and E).  But if you’re in any other family, their plan isn’t any better than the status quo, at the average.  In fact, their promise is to keep all of the cash transfers- with their various problems- the same. You have to make some pretty big assumptions about behavioural response to make the $15/day daycare payoff for average kids in average families who have subsidies or have decided to have a stay-at-home parent. The kids in families with high average daycare fees also happen to have some of the highest median household incomes (see also this).  Do they need a cap on daycare fees?  Could we do a better job in the mix of existing subsidies, credits and benefits?  The literature on childcare makes it pretty clear that the quality of the care matters more than who provides it.  If some families are choosing care outside of formal daycare, are they wrong if the quality of the care is good?

The CPC’s have, at least in rhetoric, gone to great pains to present themselves as agnostics on the daycare issue.  They left the income-tested Canada Child Tax Benefit alone and introduced a flat Universal Child Care Benefit that is, notionally, blind to whether you have one or two parents and use daycare or not. And they introduced the Family Tax Cut for couples with kids that, by design, prioritizes some family types over others. I’ve seen criticism that the CPC’s overall approach is designed to favour single-earner couples where one parent (read: mom) stays home. Well, if so, it’s not reaching the “average” families in that category.  They actually don’t do as well as the 2-earner couples because the they can’t stack quite as many benefits on top of each other.  In fact, the CPC trend line is actually sort of U-shaped when looked at on a per-child basis (not shown)–the stacking of various cash transfers and credits does much more per kid in Family E (with $99K in earned income) than for each kid in Family G (with $54K in earned income).

The LPC package does a better job of dealing with the issue of stacking benefits by replacing 3 different programs with 1 integrated benefit. By tying the benefit to household income and not the number of parents or earners, or use of daycare, the package is also more fair and more efficient for kids in the “middle-income” for different family types. More remarkably, they’ve also managed to make it a big enough program that 6 out of the 8 family types above do better under the LPC model than under either the NDP or CPC plans. For those remaining 2 families in our model (both higher income with 2-earners), it’s also clear that it wouldn’t take much in an early learning and care platform to match the NDP’s promises.

The “middle” is a crowded place in family policy these days. All the various kids there should reasonably expect policy-makers to not judge the family they come from and to make public money go as far as possible.

[1] It’s really important to note that Statistics Canada’s definition here doesn’t include kids being raised by people other than a couple or a single parent. As near as I can tell, there are thousands of kids under 18 who live with someone other than their parent or parents. These kids probably need access to financial support too and we should be asking each of the parties how their plans work for all kids – including kids in care.

[2] Data sources are Statistics Canada CANSIM Tables111-0020, 111-0015 and 111-0022. Families F, G and H are based on a male single earner which is most common in single-earner couples. For the smaller number of single-earner couples with a female earner, the median earned income is significantly lower.

[3] Source: General Social Survey, Cycle 25. Out of pocket costs were reported for childcare users by household income. I’ve used the mode for families outside Quebec.