Born for Tax Reasons

As we wind down 2015, I thought it would be timely to revisit a topic that I first explored in my inaugural blog post, posted just over two years ago to today. That blog post talked about research that had been conducted that people time deaths to benefit from a lower tax regime. You might be interested to know that research has similarly shown that people also time births to benefit from lower taxes or higher tax benefits. What, you say?! Before you start thinking that people are making plans nine months ago, what we are really talking about is manipulating the delivery date.

The seminal paper in this area was by Dickert-Conlin and Chandra (1999). They considered the U.S. child tax benefit system that granted a whole year of tax relief to an individual or family that had a child in that tax year, even if the child was born on 31 December. The authors found that such a system provided incentives for more children to be born in the last week of December rather than the first week of January; namely that a $50 increase in benefits led to a 1.4 percentage point increase in the probability of a December birth. Using a more comprehensive data source, LaLumia, Sallee, and Turner (2015), however, found a smaller magnitude: that a $1,000 increase in benefits was necessary to incite the same behavourial response as reported by Dickert-Conlin and Chandra (1999). A similar study was conducted in Japan by Kureishi and Wakabayahsi (2008) that similarly found that a system that grants a year’s worth of tax relief to parents regardless of when in the tax year a child was born increased the probability of a December birth.

These are interesting results that likely also applied to Canada prior to 1993. In 1944, Canada introduced the Family Allowance Act which provided family allowances to all children up to age 16, provided they were attending school. This is considered to be Canada’s first universal welfare program and while the payments were provided monthly, payments were based on the age of the child and not the birth month. That is, if your child was born on December 31, you got a whole year of payments for that child. It was not until 1993 with the advent of the Child Tax Benefit that the payments were calculated on a monthly basis and were means tested.

Similar behavioural responses have also been found related to the parental benefit system and the school start date system. Gans and Leigh (2009) studied the effect of a $3,000 Australian baby bonus that applied to children born on or after 1 July 2004, and announced on May 11, 2004. The authors found a significant increase in the number of births that occurred on 1 July 2004, so much so that more children were born on that day than on any other single date in the past 30 years. Neugart and Ohlsson (2013) found that an increase in parental benefits that took effect on 1 January 2007 in Germany caused births to be shifted from December 2006 to January 2007, notably a 6 percentage point increased probability of giving birth in the first week of January over the last week of December. Finally, Dickert-Conlin and Elder (2010) examine whether school eligibility cutoff dates shift births to just prior to the legislated cut-off date, arguably as a way to reduce child care costs. They found no evidence that this occurs, a finding which is unsurprising given the offsetting desire to push a birth to just after the legislated cut-off date as a way to ensure the child has a physical, mental, and maturity advantage over their child’s classmates.

These findings show that public policy can have unintended consequences on the timing of births. However, because we are only considering a change in births over a range of a few days, there are unlikely to be any follow on concerns related to birth seasonality and health, school readiness, or related issues. That said, to the degree that these shifts are caused by unnecessarily inducing births or using medical interventions to delay births, there may be increased risk of maternal and fetus health complications, something that is of great concern and why public policy makers need to account for such unintended consequences when designing supports for families with children.


Land Transfer/Registration Levies in Canada

The province of Ontario announced today that it would not be allowing all municipalities in the province to charge a land transfer tax, allowing that privilege only for the City of Toronto. The announcement provides an ideal opportunity to consider provincial land transfer taxes and fees across Canada, as they vary widely.

First, the only province in Canada that devolves the full responsibility for levying land transfer taxes (called deed taxes) to its municipalities is Nova Scotia. Municipalities in NS can set their own rates, so long as they are at or below the cap of 1.5%. Here is a link that outlines the rates across municipalities in NS, and you will see a number of municipalities do levy the maximum rate allowed.

Second, only the City of Toronto in Ontario has been given authority by the province to levy a land transfer tax. Unlike in NS where the municipal transfer tax is fully devolved, in Toronto the transfer tax charged by the City is in addition to the provincial tax.

Third, the form of the tax/fee varies across the provinces. In Alberta, Saskatchewan, and Newfoundland the levies are not land transfer taxes per se, but rather a registration fee/tax. In the case of Alberta, there is a separate fee/tax for registering mortgages. However, these so called fees all take the form of a rate that varies with the value of the property. Since the cost of registering a property does not vary by its value, the variable portion is by definition a tax and not a fee. In the remaining provinces, they are officially referred to as land transfer taxes.

Fourth, the tax rate that is applied to the property value is sometime progressive, but more often than not a flat tax. The rates are progressive in BC, MB, On, and Quebec. Flat land transfer/registration taxes exist in Alberta, Saskatchewan, New Brunswick, Nova Scotia, PEI, and, for the most part, Newfoundland (NFLD exempts the first $500 in property value). The rates vary widely, from $1 per $5,000 of assessed value to 2% of assessed value.

Fifth, most provinces apply the tax to the fair market value of the property, interpreted as the purchase price of the property, but some provinces have a caveat that allows them to use the greater of the purchase price or the assessed value. Of course, fair market value of property varies widely within and across provinces and the higher the value of the property, the greater the tax.

This variation leads to the burden of land transfer/registration taxes differing immensely across the provinces. One of the best ways, I think, to gauge the burden is to look at the fees/taxes due based on the average value of residential properties in the province. I obtained the average value of residential properties from CREA and applied the tax/fee to these averages (I assume that a mortgage was also registered in those provinces that levy a separate fee to register a mortgage). The results are listed below:

Land Transfer/Registration
Province Average Value Fee/Tax Portion of
Assuming Mortgage Value
BC 667,480.00 11,349.60 1.700%
AB 384,381.00 254.00 0.066%
SASK 292,949.00 878.85 0.300%
MB 264,739.00 2,944.78 1.112%
ON 471,654.00 5,549.81 1.177%
Toronto 630,876.00 13,892.33 2.202%
QUE 275,701.00 2,635.52 0.956%
NL 258,459.00 2,263.67 0.876%
NB 161,338.00 806.69 0.500%
NS (Max Rate) 207,657.00 3,114.86 1.500%
PEI 163,236.00 1,632.36 1.000%


You can see that among the provinces, the burden of the transfer tax in BC is the highest based on average value; however, overall, the burden is the highest in Toronto due to the presence of the combined provincial-municipal land transfer tax. The provinces with the lowest burden (AB and Sask) are those purporting to have only a registration ‘fee’ and not a transfer tax per se. Despite the varying burden, it is clear that the presence of land transfer taxes do not generally inhibit homeownership.

What is the effect of such high land transfer taxes in Toronto and BC? The CD Howe Institute has studied the effect of the City of Toronto transfer tax and the results were generally not positive. Overall, the transfer tax has constrained mobility, lead to increased administrative costs, and pushed home buyers to buy just outside of the municipal borders. Instead, the report suggests that a higher property tax rate would be better than the transfer tax. That said, as the city is facing a $23 billion infrastructure shortfall, it is unlikely the tax will be lowered or disappear any time soon. In BC, the high burden has led some to suggest that the tax constrains the ability for BC employers to attract and retain skilled workers, despite the provinces lower income taxes. And while the province of BC has indicated that it is willing to tinker with the land transfer tax, it has also indicated that it is unwilling to forgo much of the $1 billion in annual revenue it obtains from the tax. In BC, the cost of home ownership is often semi-jokingly referred to as the sunshine tax; that price one pays to live in a mild climate.

What is, however, clear from looking at these rates is that provincial policy regarding land registration/transfer taxes is not consistent. BC is a definite outlier amongst its peers, but then so too is Alberta. While the previous Alberta government had proposed to increase the land registration levy, the new Notley government will not follow through on that proposal. Given the table above, it would appear that there is room for Alberta to look to the land registration system as a way to obtain much needed revenues; however, with the effect of the oil price collapse already working its way through to the housing market, the timing perhaps is not ideal for such change.