Patrick Grady and Herbert Grubel
Fact Checking the CBC's "Fact Checker" on the Fiscal Cost of Immigration
August 5, 2019
In "fact checking" Maxime Bernier’s speech on immigration (Bernier, 2019), the CBC’s Jonathon Gatehouse (2019) played fast and loose with the "facts" in pronouncing Bernier's claims about the costs of immigration as "patently false" (Gatehouse, 2019).
To support his claim that "Immigration is actually very costly for governments," Bernier cited our 2011 peer-reviewed study published by the Fraser Institute (Grubel & Grady, 2011). He could have cited our two subsequent peer-reviewed studies (2012; 2015), which provide updates, but he didn’t. However, this really isn’t a problem as, even though our methodology has been slightly revised in response to criticisms received (Javdani and Pendakur, 2011), and applied using more recent data as it became available, our overall finding that there is a substantial net fiscal cost resulting from recent immigration (since 1985) remains essentially the same as claimed by Bernier. More specifically, while the annual net fiscal transfer to recent immigrants calculated in our updated study was slightly lower at $5,329 per capita in 2014, when applied to the number of recent immigrants, it still yields an even greater total fiscal cost of $27 to $35 billion in 2014 because of the additional number of immigrants arriving after 2006, the end of the period covered by our earlier study.
The methodology of our study involves the utilization of data on income, earnings, taxes, and immigration from the Census to allocate Statistics Canada data on government revenues and expenditures between recent immigrants and other Canadians. The reason recent immigrants were chosen rather than all immigrants was that there was a major shift in the source of immigrants in the mid-to-late 1980s away from traditional source countries to the Third World. This has been accompanied by a rather dramatic decline in the relative performance of immigrants in the labour market that has given rise to the growing concerns about the rising cost of immigration.
It should be obvious (but it apparently is not to Mr. Gatehouse) that any determination of the fiscal cost of immigration to governments required to reliably "fact check" assertions on the fiscal cost of immigration needs to rely on information on government revenues and expenditures and taxes paid and not just on the labour market performance of immigrants. The absence of any information on immigrant-related government revenue and expenditure is the most serious of the many flaws of Mr. Gatehouse’s "fact check."
Mr. Gatehouse was quick to dismiss the carefully-crafted results of our study and to turn to a consideration of the most recent labour market and census data. While this data does a good job of showing the extent to which Canadian economic growth has come to depend on immigration because of the slowdown in the natural growth of the Canadian population, it sheds little light on the growth of Canadian incomes per capita. And most importantly for this "fact check," it tells us nothing of the fiscal cost of immigration.
Mr. Gatehouse does present some interesting data from a Statistics Canada study of immigrant labour market trends (Yssaad and Fields, 2018). Cherry picking to make his case, he emphasizes the narrowing in the unemployment rate gap between recent immigrants and native-born Canadians for men. But he fails to point out the persistence of high gaps for women.
The fundamental problem with labour market data taken by itself is that it doesn't tell us anything about income earned from employment or other sources. High employment rates for recent immigrants combined with low unemployment rates do not necessarily lead to high incomes if a disproportionate share of the jobs are at lower wages. It is incomes earned and taxes paid that are most relevant to the question of the cost to the Government of immigration and a "fact check" that does not look at this is not worthy of the name.
Mr. Gatehouse gets closer to the question at hand in examining data from another Statistics Canada study on the income and mobility of immigrants that shows the highest entry wage for immigrants occurring in 2017 and consistent wage increases for the 2006 cohort of immigrants over the last ten years (Statistics Canada, 2018). Again, interesting information, but by itself not capable of providing a "fact check."
Mr. Gatehouse finally does get around to looking at the 2016 Census data (Statistics Canada, 2019), which, if used correctly, can provide part of the answer to the question (but not the full answer as information on Government revenue and expenditures is also required). He notes the large $13,941 gap in average employment earnings of someone who has been in the country for under five years compared to the average native-born Canadian. He makes much of the decline in the gap over time with the difference shrinking to just under $4,900 for those in the county for a decade, and then actually reversing to $5,500 for those who have been in Canada for 25 years or more. The problem with this, however, is that by comparing the employment income of an older group of workers with the average of all non-immigrants young and old it confuses increasing earnings over a career as a result of aging and greater experience with gap closing and ignores the continuing inflow of new underperforming immigrants. This is a novice mistake often made by people unfamiliar with cross-sectional data like that from the Census. Someone in the country 25 years or more obviously needs to be compared with an older group of people higher up the salary ladder than the average. In addition, Mr. Gatehouse only considers employment income and not total income, where there is a larger gap for immigrants who don't have as much non-wage income, especially when they haven't been in the country long. Remember it is total income that is the base of Canada’s most important tax, the unloved and highly progressive personal income tax. And it is the total income of all the immigrants not just those who have been in the country more than 25 years. The average total income of all immigrants is only 81.2 per cent of non-immigrants according to the 2016 Census data hyperlinked in Mr. Gatehouse's article.
Since our most recent study used data from the 2011 Census and National Household Survey, it is instructive to consider some of the data from the more recent Census tabulation cited by Gatehouse to see if the income gaps observed in our earlier studies have persisted in the 2016 Census. For this purpose, we consider the data for the 25 to 54 "core" age group, which is more comparable than data for all ages and is presented in Table 1 (although this age group is too large for more exacting analysis). It shows that average total income for immigrants in this age group who arrived almost a decade or more ago was only 76 per cent of non-immigrants. And even immigrants who had been in the country for almost 20 to 30 years only had average total income of 83.6 per cent of non-immigrants. This means that, since, at least, income taxes are progressive, the taxes paid by recent immigrants will be an even lower percentage of that paid by non-immigrants than the income figures indicate. Incidentally, the employment earning gaps for all ages for immigrants landing from 1991 to 2000 and from 2001 to 2010 (not shown on the table) is similar, but slightly smaller. As the reader can see in Table 1, Mr. Gatehouse’s claim the vast majority of immigrants are “earning, almost exactly as much as other Canadians” is belied by the data.
|Average total income ($)||58,159||47,202||48,650||44,177||35,859|
|Average total income
(% of Non-Immigrants)
Source: Census 2016.
The main "fact" so glibly declared false in Mr. Gatehouse's "fact check" is the question that Mr. Bernier raises, namely: "Are Canadians happy to subsidize 74% of our current immigrants?" Mr. Bernier’s "74%" presumably comes from the proportion of immigrants that aren't classified as principal applicants under economic class immigration programs mentioned in his speech. Granted that doesn't necessarily prove that these immigrants won't do well, but it does indicate that those immigrants were not selected by the Government based on their presumed economic contribution to the country so if they do end up doing well it wasn’t because they met carefully crafted selection criteria established by the Government to ensure the good performance of the immigrants selected.
The real answer to the question of the fiscal contribution of all immigrants requires extensive empirical analysis of the type we did in our continuing studies over more than a decade, starting with Grubel (2005). It can't be assumed that family and refugee class immigrants are costly to the Government, just as it can't be assumed that economic class immigrants are not. It is necessary to look at the actual data.
Before proceeding further, it is necessary to specify what is meant by "subsidized." This is a non-precise term that could have many meanings and needs to be more precisely defined to be made more operational. The most reasonable interpretation in the current context of the cost of immigration to the Government is that a "subsidized immigrant" is one who pays less to the Government than he or she gets back in benefits from expenditure programs provided by the Government. This is not a simple matter of counting as "subsidized" immigrants who receive a check from the Government clearly marked "immigration subsidy." No such official subsidy exists. Rather it requires complex calculations and estimates of Government revenues and expenditures paid by and to recent immigrants and others. It also involves complicated questions of tax and expenditure incidence about which economists can and do disagree. And most importantly to recognize, it is not a simple "fact" that can be easily checked by a CBC "fact checker," who chooses not look at all the relevant data, and who is not professionally trained to carry out such analysis.
A key feature of Canada's system of government revenues is that the contributions and taxes paid by individuals rises with income. The more people earn, the more they pay in taxes and other contributions to Government. The personal income tax in Canada is progressive and the amount paid by individuals increases more rapidly than income. In fact, a recent study published by the Fraser Institute estimates that over half (56 per cent) of taxes in Canada is paid by 20 per cent of households (Lammam, MacIntyre & Palacios, 2017). The steep progressivity of the Canadian personal income tax is documented in another Fraser Institute study (Milligan, 2015, p.57-59). It is these taxes and other revenues, including lucrative resource revenues, that finance Government expenditures that provide benefits to all Canadians.
In contrast to Government revenues, benefits from Government expenditures tend to be provided more equally on a per-capita basis to all Canadian residents. For instance, health and education costs do not necessarily go up with the income of the recipient. And they are available to all new immigrants on the same basis as they are for other Canadians. The availability of the benefits provided by the Canadian welfare state is one of the things that makes Canada such an attractive destination for immigrants.
The combination of relatively fixed per-capita benefits from Government expenditures combined with escalating taxes and other Government revenues as income climbs results in a transfer from those at the upper end of the income scale to other Canadians. When received by recent immigrants, this fiscal transfer can be considered the net fiscal cost of immigration. This is what we estimated in our studies.
Based on this methodology and the generally known fact documented in the Census that recent immigrants earn substantially less than other Canadians, it is puzzling to see how Mr. Gatehouse could so cavalierly dismiss Mr. Bernier’s question (claim?) as false. He doesn’t seem to understand that the more immigrants who enter the country and do less well economically than the existing Canadian population and consequently receive more in Government benefits than they pay in taxes, the higher the cost of immigration to the Government, and the greater the income redistribution through the Government that takes place from the existing Canadian population to new immigrants. It’s not enough for new immigrants to all get jobs as argued by Mr. Gatehouse, as desirable as that certainly is, they must also make enough money from these jobs to pay their share of maintaining what is a very generous and expensive welfare state. Those that don't at least pull their own weight are in a very real sense being "subsidized" by other Canadians whether the CBC calls it so or not.
Mr. Gatehouse and the CBC would be well advised to limit their "fact checking" to more simple "facts" that are within its areas of competence and that can be easily verified rather than trying to make judgements on complex technical and analytical issues that they so obviously don’t fully understand. To do otherwise is to abuse their position of trust as the nation's respected and supposedly unbiased reporter of news, especially now in what is a politically charged and hotly contested election climate.
Bernier, Maxime (2019) "The People’s Party of Canada Position on Immigration and Multiculturalism," July 24.
Gatehouse, Jonathon (2019) "Politics·Fact Check: Maxime Bernier's false claim about Canada's 'subsidized' immigrants," CBC, July 26.
Grady, Patrick and Herbert Grubel (2015) "Immigration and the Welfare State Revisited: Fiscal Transfers to Immigrants in Canada in 2014," Fraser Institute, November.
Grubel, Herbert (2005) "Immigration and the Welfare State in Canada: Growing Conflicts, Constructive Solutions," Public Policy Sources 84. Fraser Institute, September.
Grubel, Herbert and Patrick Grady (2011). "Immigration and the Welfare State 2011," Fraser Institute, Study in Immigration and Refugee Policy, May.
Grubel, Herbert and Patrick Grady (2012). "Fiscal Transfers to Immigrants in Canada: Responding to Critics and a Revised Estimate," Fraser Institute, Study in Immigration and Refugee Policy, March.
Javdani, Mohen and Krishna Pendakur (2011). "Fiscal Transfers to Immigrants in Canada," Metropolis British Columbia Working Paper No. 11-08.
Charles Lammam, Hugh MacIntyre and Milagros Palacios (2017) "Measuring the Distribution of Taxes in Canada: Do the Rich Pay Their 'Fair Share'?" Fraser Institute, November 13.
Milligan, Kevin (2015) "The Progressivity of the Canadian Personal Income Tax," Paper Prepared for University of Calgary School of Public Policy and Canadian Tax Foundation Conference "Reform of the Personal Income Tax in Canada," held April 15-16, 2015.
Statistics Canada (2018) "Income and mobility of immigrants, 2016," The Daily, December 10.
Statistics Canada (2019) Data tables, 2016, Census, Immigrant Status and Period of Immigration (10), Income Statistics (17), Age (10) and Sex (3) for the Population Aged 15 Years and Over in Private Households of Canada, Provinces and Territories, Census Metropolitan Areas and Census Agglomerations, 2016 Census - 25% Sample Data.
Yssaad, Lahouaria and Andrew Fields (2018) "The Canadian Immigrant Labour Market: Recent Trends from 2006 to 2017," The Immigrant Labour Force Analysis Series, 71-606X, December 18.