Worthwhile Canadian Initiative

No, your kids don't want your stuff

Dining room sets that cost thousands of dollars new can be picked up for a few hundred dollars (or less) on Kijiji or Craigslist. Here's a quote from typical ad: "Solid wood in Excellent condition. Rarely used. I am moving to a smaller house and it needs to go. Price is negotiable." 

It's part of a larger trend. While the price of household furnishings and equipment has doubled over the past 45 years, the price of shelter has quadrupled - and that's just average shelter costs. These numbers do not not take into account the difference between the shelter costs faced by people entering the housing market now and those faced by people who bought houses back in the 1970s, 80s or 90s.

 

So - a plea to my fellow baby boomers... Your kids might admire your wooden salad bowls. They might appreciate the fact that you carefully preserved the diorama they created in first grade. But they don't want your stuff. They can't afford to buy a home with space to store it in. And if they want dining tables or tchotchkes, they can pick them up for next to nothing elsewhere.

As the kids would say: let it go.

From tax policy to tax politics.

Notes prepared for a Session in Honour of Robin Boadway II: Emerging Issues in Tax and Transfer Policy, held at the Canadian Economics Association meetings in Toronto, June 1, 2024.

The fundamental principles of optimal tax analysis have not changed greatly over the past 50 years. The constraints faced by tax policy makers have.

The sources of taxable income have shifted. In 1976, the earliest year for which data is available, men's wage and salary income made up almost two-thirds of "market income" - income coming from employment, self-employment, private pensions, and investments. Now it accounts for less than 50 percent. 

Women's wage and salary incomes now contribute substantially more to the tax base than they did in the 1970s. This matters to the extent that it changes the elasticity of labour income. Generally speaking, female labour supply is more elastic than male labour supply (for recent evidence see Michael Keene (gated) or this 2008 ungated article by Meghir and Phillips) The more elastic an income base, the greater the economic costs - in terms of distortions to people's behaviour, reductions in well-being, deadweight loss - of taxing it. Hence the changing composition of the income tax base potentially impacts the ability of governments to raise income tax revenue, and the relative desirability of capital v. income taxation. 

 

Population aging is one reason why men's male and salary income now accounts for a smaller share of the income tax base: the decline in overall (male plus female) wage and salary income shown above reflects an increase in retirement income share.  But what does a decline in the number of workers relative to the number of retirees mean for tax policy?

Generally speaking, when people get older, they spend more than they earn. Thus it is possible to partially offset the fiscal impacts of population aging by taxing spending more - by, for example, increasing GST/HST/other sales taxes - and taxing income less (see, for example, Achou et al ungated here).  But while the right tax policy choices might be fairly clear, how to implement them is less so. For a government, increasing the GST is political suicide. The big tax policy question is not "what policies should be adopted?" but rather "how can we build a political consensus around good tax policies?"

The tax treatment of retirement savings raises similar political challenges. Registered Retirement Savings Plan (RRSP) contribution room was expanded in the 1990s and 2000s, partly in anticipation of demographic change. In 2002, Marcel Mérette optimistically wrote, " the deferral of tax payments through [RRSPs and similar programs] nicely offsets the spending pressure arising from health care and public pensions during a demographic transition toward an aging population."

Today Canadians are allowed to contribute to RRSPs until age 71, after which point the funds must be converted into a Registered Retirement Investment Fund (RRIF) and gradually withdrawn - triggering tax liabilities. There's only one problem: people don't like drawing down their retirement savings, and they especially don't like paying tax on those withdrawals. Evidence from the US here and also here finds that, given a choice, a substantial number of people would make less than the required withdrawals from their registered retirements accounts (IRAs in the US, RRSPs/RRIFs in Canada). There is a never-ending stream of proposals to reduce the taxation of deferred savings by, for example, raising the age at which an RRSP must be converted to a RRIF to 75, eliminating mandatory RRIF withdrawals altogether, or exempting the first $160,000 of RRIF savings from withdrawal rules. 

Good Chart Checklist

Note: this was prepared for my ECON 3403 students, and is a list of all of the mistakes I commonly see in student charts. Please add your suggestions for things to add to this list in the comments - or steal, modify, and use for your own purposes! 

Good Chart Check List:

Is my chart self-explanatory? Does it specify

• Who: The population the chart shows (e.g. eligible voters, Facebooks users, people over the age of 12, Canadian citizens, etc)
• Where: The geographic area the chart applies to (Ottawa, Canada, G7, G20, OECD)
• When the data was collected
• The data source (e.g. Statistics Canada Table 12-345-67, World Bank). Optional: give name of survey or statistical programme data came from.

Am I using an appropriate chart type?

• Line chart to show time trends (typically 4+ observations – for just two years of data use a bar chart).
• Bar chart to compare groups of things, e.g. countries, age groups, genders, occupations, industries, provinces
• Pie charts are rarely the appropriate chart type. Use pie charts when your only concern is distribution; you only want to know how each group’s share compares to other group’s shares. If you must use a pie chart, consider using the “doughnut” option.

Are my axes clearly labelled? Is it clear

• What each axis represents Exceptions:
  • do not have to label a “year” axis, it is usually obvious that 2005, 2010, 2015, 2020 etc are years.
  • If the chart title makes it clear what the graph shows, don’t have to label the axis as well.
• What units I am using (miles, km, tonnes, millions of tonnes, Canadian dollars, 2020 (constant) dollars, nominal (not inflation adjusted) dollars, etc)

Have I put my data into perspective?

• For data showing trends over time: are dollar amounts adjusted for inflation?
• For comparisons across provinces or across countries: am I showing numbers as a percentage of the population, to adjust for population size?
• For data focussing on a particular population subgroup: have I included data for other groups to put the numbers into perspective? E.g. for a chart showing trends in 12-15 year old’s mental health, consider including mental health trends for the entire population to put the numbers into perspective. For a chart showing poverty rates for elderly, include poverty rates for the non-elderly for comparison purposes.
• For data showing trends over time: have I tried the “year 1=100” approach – i.e. divide by the year 1 value*100?

Am I being honest?

• Does my vertical (y) axis start at zero?
  • Exception: sometimes, if you’re trying to highlight trends, it’s o.k. to have a line chart that doesn’t start at zero. But be very careful! For a bar chart, the vertical (y) axis must ALWAYS start at zero!
• Am I selecting my starting year carefully to show the results I want to show? (e.g. I am I showing stock market growth starting at 2009, when the market was at very low levels, instead of at 2007?).
• Am I distorting my results using fancy graphics, especially 3-dimensional shapes?
• Are the numbers on my chart accurate?
• Have I created a combo-chart, and manipulated my axes so that my series look closely related, when they really aren’t?

Is my chart accessible?

• Can someone who is colour-blind read my chart?
  • Am I using a colour-blind friendly colour scheme, e.g. a monochromatic scheme, or a scheme designed to be colour-blind friendly?
  • Am I using dashed lines, labels, or other tools to make it easier to tell my series apart? See here for tips on how to make a graph colour-blind friendly: https://towardsdatascience.com/two-simple-steps-to-create-colorblind-friendly-data-visualizations-2ed781a167ec
• Can someone who has trouble reading small print read my chart?
  • Flip a bar chart on its side (so the bars are horizontal, not vertical). That creates room for big-enough-to-read labels!
  • Consider putting legend labels on the chart itself, rather than to the side (requires advanced Excel skills)
• Does my chart contain any unnecessary information or other clutter?
  • Avoid cluttering up a graph with too many years worth of data. Zoom in on one year and compare population groups (e.g. age, sex, industry, occupation). Alternatively, download 5+ years of data and graph a time trend (challenge: can you explain the trend?).  Statistics Canada data tables often include 3 years of data by default. However a good graph rarely includes just three years of data (exception: the years 2019, 2020, 2021).
  • Only report 2 or 3 digits worth of data – e.g. report data as 1.2 million rather than 1,234,567.
  • Cut any unnecessary lines, words or graphics
• Am I using the right words to describe wages and incomes?
  • Wage rate=$ earned per hour worked
  • Wages=total amount earned. Also known as earnings
  • Income=earnings +other sources of income.
  • For income data, try to be as precise as possible.
    • Is the data pre-tax income or after-tax? If after-tax income, does it include benefits like Canada Child Benefit?
    • Individual income or family income or household income? If family or household income, whose income is counted? (e.g. if a 25 year old is living with a parent, whose income is included in the 25 year old’s “family income”?) Check to see if the income is “census family income” or “economic family income”.

Do the self-employed opt out of the Canada Pension Plan?

In 2017, anyone with earnings was required to pay $4.95 in Canada Pension Plan contributions for every $100 of earnings (up to a maximum earnings threshold of $51,800 - rates here). For the employed, their $4.95 CPP/QPP contribution was matched by an equal contribution by their employer. However self-employed individuals are required to pay the entire employer+employee contribution themselves, amounting to $9.90 for every $100 of earnings.

Now the self-employed have an out. They can incorporate, and pay themselves in dividends rather than salary. Since they are not required to pay CPP premiums on dividends, they can potentially avoid paying into the Canada Pension Plan. 

But do they?

There are both advantages and disadvantages of opting out of CPP - sure, you don't have to pay CPP premiums, but you don't get CPP benefits either. And CPP offers some attractive benefits, for example, it provides disability insurance, and is inflation protected. Moreover, there are other reasons why the self-employed might not want to minimize their employment income. Registered retirement savings plan contribution room is tied to employment income, and financial institutions may ask for proof of employment income before they will issue a mortgage. So ultimately it's an empirical question - do the self-employed opt out of the CPP, or don't they?

To some extent, this question can be answered with the Canada Revenue Agency's Income Taxation Statistics:

The above table shows average CPP or QPP premiums paid per taxfiler. The taxfilers are categorized according to their main source of income. So, for example, the "Employment" category includes all taxfilers whose main source of income was employment, even if those individuals had some self-employment income. Some groups of people, such as individuals whose main source of income was pension income, have been excluded. 

It does seem that individuals with professional income pay quite a bit less in Canada Pension Plan contributions than one would expect based on the total income assessed per tax filer. However there just aren't that many of those tax filers. The typical self-employed person is not a doctor or a lawyer or other professional. They're a small business owner or gig economy worker - someone with proprietorship or partnership income - and they're paying CPP contributions at a somewhat lower, but not massively lower, rate than employed individuals. (They're also, in many cases, supplementing their self-employment income with employment income, and pay into the CPP both as an employee and as a self-employed individual). Now it could be that these small business owners are working for cash, or making highly aggressive use of home office and other deductions, and that their total assessed income understates their actual income. But that's another matter entirely.

The big question, however, is how many of those taxpayers whose primary source of income is "investment" are self-employed individuals paying themselves in dividends? It's impossible to answer this question directly with the aggregate CRA data. However we can get at the issue directly by using information about the Canadian labour force. According to Statistics Canada Labour Force Survey data, there were 15.8 million employed workers in Canada - a number that almost exactly matches the number in the table above - and 2.9 million self-employed (here). Even if we add the 182,750 people with fishing and farming income to the ranks of the self-employed in the table above, there are still a significant number of self-employed individuals unaccounted for. That suggests there are indeed a significant number of self-employed individuals - possibly over a million taxpayers - who are not appearing as professionals, proprietors, and so on, and not making significant CPP contributions.

 

Click here to download Self employed cpp calculations

Presidents: do you get what you pay for?

In 1915, Woodrow Wilson earned $75,000 per year, or $1.9 million in 2018 dollars, for serving as President of the United States. The current incumbent of that office receives only a fraction of that amount: $400,000 annually.

Congress has increased the presidential salary three times in the past 100 years: raising it to $100,000 in 1949, to $200,000 in 1979, and to $400,000 in 2001. These rises appear as sharp spikes in the graph below. After each one of these increases, the real value of the presidential salary has been gradually eroded by inflation, except during the deflationary 1930s. 

Can the Great Barrington proposal save the economy?

The Great Barrington Declaration argues against universal lock-downs:

Those who are not vulnerable should immediately be allowed to resume life as normal. Simple hygiene measures, such as hand washing and staying home when sick should be practiced by everyone to reduce the herd immunity threshold. Schools and universities should be open for in-person teaching. Extracurricular activities, such as sports, should be resumed. Young low-risk adults should work normally, rather than from home. Restaurants and other businesses should open. Arts, music, sport and other cultural activities should resume. People who are more at risk may participate if they wish, while society as a whole enjoys the protection conferred upon the vulnerable by those who have built up herd immunity.

Setting aside the epidemiological merits of the Declaration, the question remains: how much of an economic impact would a targeted easing up of restrictions have?

To answer that question, we need to figure out how much spending power is in the hands of "young low-risk adults", and how much is in the hands of "people who are more at risk." COVID-19 risk depends upon a variety of factors, including sex, blood-type, body mass index, and so on. However age appears to be a key risk factor - 40 to 49 year olds in the US are 10 times more likely to die from COVID than people between 18 and 29. The odds of COVID death for 50 to 64 year olds are 30 times the odds for someone between the ages of 18 and 29 (source). So, while age is not a perfect measure of COVID risk, we can use it to get a rough sense of how much of the purchasing power in our economy is in the hands of people who are vulnerable to COVID.

There are a couple of different data sets that can be used to estimate the distribution of after-tax income by age in Canada. The first is the Canada Revenue Agency's Taxation Statistics. These report the total amount of income declared by every Canadian taxfiler, broken down into five year age categories, as well as the taxes paid. They are not a perfect measure of after-tax income. They only include personal income reported for income tax purposes, not corporate income, nor any incomes not declared for tax purposes. Also, the taxation statistics do not include a measure of after-tax income, so I estimated it by calculating net income minus total taxes paid. Moreover, these numbers are from 2016, so are somewhat out of date.  But they serve to make a point.

First, even if young low-risk adults can work normally, it won't make much difference to the amount of money spent on restaurants, arts, music, sports and other cultural activities, because young low-risk adults don't earn much money. Many are in school or just starting out in the labour market. Young people's wages are relatively low, and their unemployment rates are relatively high.

In 2016, the age group that controlled the most economic resources was 50 to 54 year olds: the after-tax income of all Canadians in that age group added up to a total of $114 billion. The next highest earning age group 55 to 59 year olds, raking in a collective $110 billion after taxes, followed by 45 to 49 year olds. This should come as no surprise.  These cohorts are large. In 2016, people aged 50 to 54 were the single largest age group in Canada, followed by people 55 to 59. Moreover, people in their 50s and 40s are at the peak of their earnings potential.  

The next figure presents this same information a slightly different way. It shows the percentage of total income received by people under the ages of 20, 25, 30, and so on. For example, the graph shows that people under the age of 20, who reported a total of $9 billion of after-tax income in 2016, accounted for less than 1 percent of the total after-tax income reported that year. People under the age of 30 accounted for 11 percent of the reported after-tax income. Less than half of after-tax income was received by people under the age of 50. The bulk of income was received by people 50 and over.

As has already been noted, CRA's Taxation Statistics are not a perfect measure of after-tax income, so I double-checked the analysis using the 2016 Census PUMF. Again, the same pattern holds - roughly 10% of after-tax income is in the hands of people under 30, and the age group with the greatest after-tax income is 50-54 year olds:

Again, this data is slightly dated - but the aging of the baby boomers would be expected to push the bulk of purchasing power even higher up the age distribution.

It could be argued that people in their 20s and 30s don't earn that much money, but they're the ones who really keep the economy going. They're the ones buying homes, buying cars, buying consumer durables. This point has merit, but it only serves to strengthen the argument made in this post. Canada's housing market is on fire right now. Car sales are strong, as are sales of consumer durables. It's the service sector that's hurting - plays, music, restaurants, museums, art galleries, and so on. And, generally speaking, it's middle-aged people who have the disposable income to keep these sectors going.

Perhaps if middle-aged people were allowed to assess the risks of various activities and participate if they chose, enough would opt in to make a real difference to the health of the Canadian economy. But I wouldn't count on it.

Download data used to create this post here

Pain, brioche, and the language of taxation

Ireland's Supreme Court recently ruled that the buns Subway uses in its sandwiches contain too much sugar to be considered "bread", and are thus subject to Value Added Tax (VAT). The decision lead to headlines and discussion along the lines of "Irish High Court Rules Subway’s Sandwich Bread Is Not Legally Bread" or "Ireland declares that Subway’s bread is basically cake". The emphasis was on how "confused" and "bizarre" the entire debate was, and the cost and arbitrary nature of distinctions made in tax legislation: "Having moral distinctions between foodstuffs in your tax regulations turns out to be an awfully expensive thing".

Yet the discussion of whether or not Subway buns are bread is only confused and bizarre in English, with our barren culinary language. In French, the discussion would not seem peculiar. French distinguishes between three broad categories of baked goods: unsweetened bread is "pain", sweetened bread (often made with eggs and milk) is "brioche", and cake is "gâteau." Hence the expression "Qu'ils mangent de la brioche" is (somewhat misleadingly) translated as "let them eat cake", because there is no English word for sweetened bread.

Subway's response to the ruling was that “Subway’s bread is, of course, bread.” Yet even though the English language does not distinguish between "pain"-type bread and "broiche"-type bread, the distinction does exist - and can hold up in court. Yet the fact that Subway believed they had a plausible case, and the bemusement the case generated in the English-language media reveal the extent to which language and culture influence the types of distinctions tax systems are able to make.

The national and first official language of Ireland is Irish, rather than English. I don't know whether "arán"  - the Irish word for bread - also encompasses brioche, or the extent to which various forms of bread are distinguished in everyday speech. If you know, please leave a comment! 

Inequality and competition in a digital economy: a case study

Pokémon Go is an economy in miniature. There is exchange: players trade Pokémons and swap gifts. There is scarcity: the balls required to catch Pokémons are scarce, as is Pokémon storage capacity and other items in the game. There is production: through various activities, such as walking a Pokémon egg, or leaving a Pokémon in a gym, players can produce new Pokémon, or earn tokens. There is status seeking and signaling: high rank players parade their skill by wearing special clothing, and displaying their game level. There are even exports and imports. Pokémon Go imports dollars from the real life economy, exporting in exchange game tokens, and the enjoyment and entertainment players get from using those tokens.

Yet Pokémon Go differs from the real-world economy in two crucial respects. First, prices, the distribution of initial endowments, and all rules are determined by the game manufacturer, Niantic, which acts as an omnipotent “social planner”.  Second, anyone who does not like the rules of the game is free to leave. The possibility of exit gives power to players: because more players=more revenue, the social planner has a strong incentive to set rules that keep players happy. Thus the rules that govern the Pokémon Go economy provide insight into what type of economic rules people would choose, if they actually had power to influence the social planner’s decisions in real life. 

One of the most notable features of the Pokémon Go economy is the number of rules that limit the extent of inequality. Players earn tokens by leaving their Pokémon in gyms – up to a maximum of 50 tokens a day. Players gain status by achieving higher levels – but the game is structured so that the overwhelming majority of players are somewhere between level 30 and level 40. Players gain the ability to defeat other players in battle by powering up stronger Pokémon – but each Pokémon’s maximum combat power is capped too.

But so what? Pokémon Go is just a game, after all.

Men, women, and the end of mandatory retirement

Back in 2014, I wrote a blog post on the end of mandatory retirement for university professors. I quoted a number of men who argued that having a standard retirement age hurts women. Here's an extract from that original post:

Thomas Klassen and David Macgregor, writing in the CAUT (Canadian Association of University Teachers) Bulletin, challenged ageism in academy on the grounds that "Mandatory retirement at an arbitrary age is devastating for female faculty who often began their careers later than males and may have had interruptions to raise children."

Gender-equity is also the basis of Lloyd Spurrell and Ahmed Hussein's arguments against what they call "Rule 65":

Because of their longer life expectancy, women must contribute significantly more to their pension plan by the time they are 65, if they are to be as financially secure as a man in the same situation. Women have to contribute enough for 20 years while men have to contribute enough for only 15 years. Why should women be discriminated against in this manner?

Yes, there are older female academics who will enjoy greater financial security as a result being able to work past 65. But let's think not about anecdotes - the stories of particular men and particular women. Overall, how many of the beneficiaries from the end of mandatory retirement are men, and how many are women? Who bears the costs of the transition?

Almost 15 years after the end of mandatory retirement, evidence on who takes advantage of the opportunity to work longer is mounting up. Unsurprisingly, it's those who were advocating for the elimination of a standard retirement age - men.

The CAUT almanac, available here, gives the number of full-time university professors in Canada by age.  Here is the number of male and female full-time university teachers by in 2007-8, shortly after the standard retirement age was eliminated (rounded to the nearest 10), along with the percentage who are male:

Number of full-time university teachers by age, 2007-08
Rounded to nearest 10. Source: Calculated from 2010-11 CAUT Almanac, Table 2.10.
	50 to 54	55 to 59	60 to 64	65 to 69	70+
Male	4,060	3,950	3,880	1,180	210
Female	2,340	1,910	1,240	230	30
% male	63.4%	67.4%	75.8%	83.7%	87.5%

Unsurprisingly, in 2007-8, men pre-dominated in the oldest age groups - a legacy of the barriers women once faced in accessing full-time, tenure-track university positions.

But let's age these people 10 years, and see how many of them are still employed a decade later. If we take the 50 to 54 age cohort, and age them up 10 years, we actually have slightly more attrition among men than among female. This is not really surprising - the mortality  rate for men in their 50s is higher than the mortality rate for women. Moreover there is a big financial hit associated with retiring before 65 in a lot of pension plans, and women may not be able to afford to take that hit.

However if we look at the 55 to 59 cohort in 2007-8, and compare it to the 65 to 69 cohort in 2017-18, we see that it has become more male-dominated over time. The over-representation of men continues to increase as we move up the age distribution: in 2017-18 the 70+ cohort was 78.8% male. 

Number of full-time university teachers by age, 2017-18.
Source: CAUT Almanac
	60 to 64	65 to 69	70+
Male	3,060	1,815	1,113
Female	1,902	801	300
% male	61.7%	69.4%	78.8%

It's not hard to figure out what's going on here. People make retirement decisions by weighing up the costs and benefits of continuing to be employed. 

There are lots of reasons to expect that female academics typically benefit less than men do from continuing in academic employment past the age of 65: they're less likely to have young or university-aged children to support, they're less likely to have a younger spouse who is still working (couples tend to synchronize retirement decisions), and they're less likely to be respected by colleagues and students as wise experts and leaders. Our society doesn't value older women, and universities reflect the values held by society as a whole.

There are also reasons to expect that the costs of continuing in employment will be higher for female professors -- the opportunity cost of not being able to care for a parent, partner, or grandchild, and the costs of meeting students' expectations that their female professors will be helpful, caring and organized.

I'm pretty sure that, for me, the costs of continuing to work past 65 will far outweigh the benefits, and I'll be happy to retire. But even if I was still enjoying teaching, I still believe retiring at 65 is the right thing to do.

First, there are juniors out there who are as smart and capable as me, if not smarter. They deserve the opportunity to have an academic position, and make an impact on the world.

Second, mandatory retirement is a bulwark against the end of tenure. If universities can't get rid of professors when they're old, they'll start finding other means to shed unwanted professors - and that's unlikely to be pretty.

Thomas Klassen and David Macgregor, writing in the CAUT (Canadian Association of University Teachers) Bulletin, challenged ageism in academy on the grounds that "Mandatory retirement at an arbitrary age is devastating for female faculty who often began their careers later than males and may have had interruptions to raise children."

Gender-equity is also the basis of Lloyd Spurrell and Ahmed Hussein's arguments against what they call "Rule 65":

Because of their longer life expectancy, women must contribute significantly more to their pension plan by the time they are 65, if they are to be as financially secure as a man in the same situation. Women have to contribute enough for 20 years while men have to contribute enough for only 15 years. Why should women be discriminated against in this manner?

Yes, there are older female academics who will enjoy greater financial security as a result being able to work past 65. But let's think not about anecdotes - the stories of particular men and particular women. Overall, how many of the beneficiaries from the end of mandatory retirement are men, and how many are women? Who bears the costs of the transition?

Almost 15 years after the end of mandatory retirement, evidence on who takes advantage of the opportunity to work longer is mounting up. Unsurprisingly, it's those who were advocating for the elimination of a standard retirement age - men.

The CAUT almanac, available here, gives the number of full-time university professors in Canada by age.  Here is the number of male and female full-time university teachers by in 2007-8, shortly after the standard retirement age was eliminated (rounded to the nearest 10), along with the percentage who are male:

Number of full-time university teachers by age, 2007-08
Rounded to nearest 10. Source: Calculated from 2010-11 CAUT Almanac, Table 2.10.
	50 to 54	55 to 59	60 to 64	65 to 69	70+
Male	4,060	3,950	3,880	1,180	210
Female	2,340	1,910	1,240	230	30
% male	63.4%	67.4%	75.8%	83.7%	87.5%

Unsurprisingly, in 2007-8, men pre-dominated in the oldest age groups - a legacy of the barriers women once faced in accessing full-time, tenure-track university positions.

But let's age these people 10 years, and see how many of them are still employed a decade later. If we take the 50 to 54 age cohort, and age them up 10 years, we actually have slightly more attrition among men than among female. This is not really surprising - the mortality  rate for men in their 50s is higher than the mortality rate for women. Moreover there is a big financial hit associated with retiring before 65 in a lot of pension plans, and women may not be able to afford to take that hit.

However if we look at the 55 to 59 cohort in 2007-8, and compare it to the 65 to 69 cohort in 2017-18, we see that it has become more male-dominated over time. The over-representation of men continues to increase as we move up the age distribution: in 2017-18 the 70+ cohort was 78.8% male. 

Number of full-time university teachers by age, 2017-18.
Source: CAUT Almanac
	60 to 64	65 to 69	70+
Male	3,060	1,815	1,113
Female	1,902	801	300
% male	61.7%	69.4%	78.8%

It's not hard to figure out what's going on here. People make retirement decisions by weighing up the costs and benefits of continuing to be employed. 

There are lots of reasons to expect that female academics typically benefit less than men do from continuing in academic employment past the age of 65: they're less likely to have young or university-aged children to support, they're less likely to have a younger spouse who is still working (couples tend to synchronize retirement decisions), and they're less likely to be respected by colleagues and students as wise experts and leaders. Our society doesn't value older women, and universities reflect the values held by society as a whole.

There are also reasons to expect that the costs of continuing in employment will be higher for female professors -- the opportunity cost of not being able to care for a parent, partner, or grandchild, and the costs of meeting students' expectations that their female professors will be helpful, caring and organized.

I'm pretty sure that, for me, the costs of continuing to work past 65 will far outweigh the benefits, and I'll be happy to retire. But even if I was still enjoying teaching, I still believe retiring at 65 is the right thing to do.

First, there are juniors out there who are as smart and capable as me, if not smarter. They deserve the opportunity to have an academic position, and make an impact on the world.

Second, mandatory retirement is a bulwark against the end of tenure. If universities can't get rid of professors when they're old, they'll start finding other means to shed unwanted professors - and that's unlikely to be pretty.

Can you be confident about an economy you can't see?

Canada's economy may be on life support, but it is still hanging on. Even on streets that seem empty and shuttered, there are a few businesses getting by, or even thriving.

The bookstore on Beechwood Avenue looks closed, but every so often a customer pops by to pick up an order from the bookrack hidden in the vestibule. At Life of Pie, the neighbourhood bakery/restaurant, the tables are stacked up, and the lights dimmed - but in the back the kitchen is humming, as the staff works hard to fill a steady stream of orders. When I ordered take-out from the local  Indian restaurant, the owner sounded busy and cheerful for the first time in months. The local farmer's market has moved on-line - no more lingering over the potatoes and fighting hipsters for the last bag of arugula. All orders now have to be in by Thursday noon, paid in advance by e-transfer, and picked up in a quick, physically distanced transaction. There is economic activity going on, but much of it is invisible, and hidden at home. As for the rest - if you don't look closely, you'll miss it.

I hope that policy makers can be creative and figure out ways of opening up more local businesses. But even if they do, what will that do for consumer confidence? Can you be confident in an economy that you can't see?

The behavioural economics of the Marie Kondo method

Marie Kondo is the guru behind the best-selling Life-Changing Magic of Tidying Up and Spark Joy. For some people, like me, her method works. One possible reason for its success is that, underneath it all, there are some sound behavioural economics principles.

1. Shift the reference point

Marie Kondo recommends dividing one's possessions into five categories and tidying them a category at a time: first clothes, then books, papers, miscellaneous and finally sentimental items. Within each category, the first step in tidying is to empty all of one's cupboards, closets and drawers, and put every single item of say, clothing, into one big pile.

It's a brilliantly simple and effective nudge.

Moving one's clothing creates a new, shifted, reference point. The default or status quo position changes from "in the closet" to "on the floor." Taking each item of clothing and putting it back in the closet then involves a deliberate choice; an act of will. (Marie Kondo emphasizes that each item must be dealt with one at a time - shoving a pile of t-shirts back into the t-shirt drawer is not allowed). Once the status quo has changed to "on the floor" it's much easier to take the next step and say "off to the thrift store."

2. Counter loss aversion

Homes become cluttered and untidy when they contain too much stuff. But it's hard for people to part with things, because we are loss averse - we feel the pain of giving something up much more acutely than the thrill of getting something new. 

The Marie Kondo method creates a series of rituals to counter loss aversion. Pick up an article to see if it "sparks joy". If it does not, thank it for its service, and let it go. However hokey it seems to say an emotional farewell to a t-shirt, the act fits with standard psychological advice: the best way to cope with negative emotions is to observe your feelings, acknowledge them and work with them. Allow yourself to feel sad, and the sensation will pass; suppress those emotions, and they will linger and fester.

Reading Amartya Sen on Poverty and Famines

The shut down of entire sectors of the economy in response a pandemic is akin to a crop failure. Both represent a sudden, large and unexpected decline in production. Both leave people without resources, without the wherewithall to command the necessities of life. So I am revisiting Amartya Sen's 1981 monograph Poverty and Famines (ungated), to see if his entitlement approach has relevance for today's crisis.

Sen begins with the crucial distinction between there being enough and people having enough.

In Canada there is enough toilet paper for everyone. Enough clean water. Enough food. But not everyone has enough of these things. Why not?

The key concept Sen develops in Poverty and Famines to explain deprivation is "entitlements". An entitlement is, in essence, an ownership claim. People go hungry when, for some reason, they are not entitled to food. 

Sen identifies four main sources of entitlements - trade, production, own-labour, and inheritance/transfer.

Transfer entitlements are the reason why few people starve in rich countries:

....The reason why there are no famines in the rich developed countries is not because people are generally rich on the average. Rich they certainly are when they have jobs and earn a proper wage; but for large numbers of people this condition fails to hold for long periods of time, and the exchange entitlements of their endowments in the absence of social security arrangements could provide very meagre commodity bundles indeed. With the proportion of unemployment as high as it is, say, in Britain or America today, but for the social security arrangements there would be widespread starvation and possibly a famine. What prevents that is not the high average income or wealth of the British or the general opulence of the Americans, but the guaranteed minimum values of exchange entitlements owing to the social security system. (Sen, Poverty and Famines, pp. 6-7 emphasis added)

Sen wrote these words before the Clinton-era welfare reforms, before the rise of the gig economy and independent contractors; before Thatcherism and Britain's austerity era. The "guaranteed minimum values" are not what they once were.

This is no virus for old men

Statistics Canada has just released a dataset with detailed anonymized information on all confirmed COVID-19 cases in Canada, available for download here.

Unlike many of the available trackers, the Statistics Canada data reports cases by date of onset, defined as "earliest date available from the following series: Symptom Onset Date, Specimen Collection Date, Laboratory Testing Date, Date reported to the province/territory or Date reported to Public Health Agency of Canada." Defined that way, the case numbers start to ramp up in the second week of March. The decline in late March almost certainly reflects the time lag between symptom onset and confirmed COVID-19 diagnosis, rather than any peaking of the curve. 

Using the data, it is also possible to estimate hospitalization probabilities, that is, the predicted likelihood that someone with a confirmed COVID-19 diagnosis will end up hospitalized. This is no virus for old men - roughly 70 percent of men over the age of 80 diagnosed with COVID-19 end up hospitalized [corrected]. This diagram does, of course, overstate the true hospitalization probabilities, as people with mild cases of COVID-19 are unlikely to be tested - or even eligible for testing - and also unlikely to be hospitalized.

Regional disparities in hospital bed access across Canada

Thank you to Kayle Hatt and Kevin Andrew for generating some of the numbers and charts behind this post, and for helpful suggestions.

The number of acute care hospital beds per capita is an imperfect measure of a health care system's ability to respond to the COVID-19 pandemic. After all, beds are easy to acquire, as are hotels or similar buildings to put beds in. It is ventilators, IV drips, heart monitors, masks, gloves, and, above all, skilled health professionals that are the crucial resource constraints in this crisis. However if most hospitals operate with more or less the same ratio of nurses/beds, IV drips/beds, ventilators/beds and so on, looking at the number of beds per capita is a pretty good way to compare different countries', or regions', health care capacity.

In my last post and this piece for IRPP I discussed Canada's low number of beds relative to other OECD countries, and interprovincial disparities in bed numbers. In this post, I drill down to the health region level, show just how much hospital bed numbers differ across regions within the same province. This gives a useful indicator of Canada's regional disparities in health care funding and access to health care - and thus raises questions about the overall structure of health care financing in this country, and the allocation of health care resources.

The most dramatic regional differences in beds per thousand people can be found in Ontario:

How prepared is your hospital for COVID-19?

I've written a blog post for the Institute for Research on Public Policy (IRPP).  It begins:

One widely touted response to the COVID-19 pandemic is to “flatten the curve” — to spread COVID-19 infections over time, so that the medical system can cope with them. Yet a cold hard look at the numbers suggests our hospitals cannot cope with the most flattened of curves. Indeed, they cannot cope with any kind of curve at all.

Canada has 1.95 acute care hospital beds per 1,000 people, fewer than any other OECD country but Mexico. (Italy, which has been overwhelmed by this coronavirus, has 34 percent more beds per capita than we do: 2.62 per 1,000 people.) Nationwide, the occupancy rate for Canada’s hospital beds is over 90 percent. (To put that number into perspective, the occupancy rate for US hospital beds is 64 percent.) Canada’s bed numbers and occupancy rates together imply that the “spare capacity” in our health care system is, at best, around 2 acute care beds per 10,000 people. If even a small fraction of Canadians contract COVID-19, and a non-trivial portion of those require hospitalization, our system will be overwhelmed.

How did Canada end up in this situation, and what can be done about it? (read the rest here).

 

The provincial level calculations in the post are based on Canadian Institute for Health Information (CIHI) data on the number of beds per hospital, per health region, and per province. You can download the raw data, plus a pivot table showing bed numbers at the health region and province level, here: Download Beds per health region

 

What would be really interesting is to do some calculations at the health region level, and find out which health regions are particularly short of beds. Unfortunately I've been unable to find data on the number of people per health region in Canada. Statistics Canada does produce a correspondence file that maps health regions to census districts, see here: https://www150.statcan.gc.ca/n1/pub/82-402-x/82-402-x2018001-eng.htm. Population per census district can be found here: https://www12.statcan.gc.ca/census-recensement/2016/dp-pd/hlt-fst/pd-pl/Table.cfm?Lang=Eng&T=703&S=87&O=A#, or calculated from the Census PUMF, which is can be freely downloaded here: https://www150.statcan.gc.ca/n1/en/catalogue/98M0001X (click through the links, it is actually free and downloadable. (I tried pulling numbers from the Canadian Community Health Survey, but that doesn't seem to work).

 

A project for someone with a computer and time on their hands?

Postmodern Economic Measurement: There is no number, only numbers

National income accounting distills a nation's economy into a single number: gross domestic product. GDP has been critiqued many times for its neglect of household production, environmental degradation, income distribution, and so on.  Many alternatives have been proposed, such as genuine progress indicators, happiness measures, the Human Development Index, the Better Life Index, well-being indexes, or prosperity indexes. Yet none have dethroned GDP as the king of economic indicators.

GDP's prominence is partly due to history - once a measure is widely accepted and understood, it is costly to switch. Moreover, GDP has generally been a good enough measure of the economy. It has a reasonably straightforward interpretation, and is a fairly good indicator of the economic things that matter: people's incomes, their ability to command goods and services, and whether the overall level of economic activity is expanding or contracting.

Yet structural and technological changes threaten to undermine the usefulness of GDP. The link between GDP growth and the typical person's standard of living is weakened by increased income inequality, which allows the rich to capture a disproportionate share of any increases in GDP. Technological change weakens the link between GDP and consumer welfare. GDP measures the value of a good or service by considering its cost of production. But in a digital economy, the marginal costs of production for many goods and services are  close to zero. Thus there is a gap between the amount that an app or game or youtube video contributes to GDP, and its contribution to consumers' well-being or satisfaction. Technological change also complicates national income accounting. For example, if Toronto company hires a freelancer in Delhi through the Tel Aviv-based website fiverr.com, how is each country's GDP affected? (HT to Armine Yalnizyan for many of these points).

Last, but by no means least, population aging could cause Canada to "turn Japanese" and enter an era of low- to no- GDP growth. In this case, governments will not wish to be judged by their GDP performance. Indeed, New Zealand, Scotland, Iceland, and now Canada have committed incorporating quality of life measurements into government decision-making and budgeting.

Does (cohort) size matter?

In the US, the portion of young men between the ages of 18 to 34 who report having at least one partner has fallen substantially in recent years (sorry for the small image size):




Charts similar to the one above have prompted talk of a sex recession. Yet worries that Millennials are killing sex (as well as napkins, diamonds, and casual dining) may be premature. This data comes from the GSS, which samples a few hundred people between the ages of 18 and 34. The 95 percent confidence intervals around these estimates, shown by the whisker bars on the chart, are very large. For example, the margin of error on the 2018 estimate for men (77.5% with at least one partner) is 7.5 percentage points.

In my submission to the annual Macleans Charts to Watch compilation here I compare the GSS numbers to those obtained from the much larger Canadian Community Health Survey. I find a smaller, but similar, downwards trend in young men's sexual activity rates north of the border. With the larger sample, I could pinpoint the downturn more precisely: it is concentrated among 20 to 24 year olds.

It does seem that these numbers are picking up something real. The question is, what?

Why do I have to collect my pension already?

Ontario's Ford government has a plan to induce professors over 71 to retire. I wrote a column in the Globe and Mail about it. Here's a sneak peak:

Nearly one in 10 Ontario university professors is over the age of 65. As of 2016, these professors were earning, on average, $184,947 a year. Moreover, because federal legislation requires all taxpayers to start drawing down their retirement savings at the age of 71, septuagenarian professors can collect a six-figure pension on top of a six-figure salary.

It sounds expensive – and it is. A report from the Higher Education Quality Council of Ontario on universities’ fiscal sustainability estimated that, if mandatory retirement was still in place, and Ontario universities had been able to replace all of their professors over 65 with junior scholars, their annual salary costs would be $89-million lower, and more than 1,200 new PhDs could have been hired.

In its April budget, the Ontario government committed itself to ensuring “a more sustainable postsecondary sector” and “employee renewal.” The details of the government’s plans are not yet known. However, a discussion paper released earlier this year reported that “the ministry is giving consideration to a policy that would … require institutions to reduce salary payments so that salary and pension payments combined are not greater than the employee’s salary prior to pension payments commencing.” Professors aged 71 and older would have their salary clawed back, giving them every incentive to retire.

The rest of the article is here. In it I talk about the various objections to the Ford government's plan, and what I see as a reasonable alternative. One thing I couldn't fit within the 800 word limit, however, is the answer to the question, "Why do I have to collect my pension already?" Why couldn't the problem of double-dipping be solved by allowing septuagenarian professors to defer their pensions. Here I explain why this will not, and should not, be an option.

About that EKOS poll

A recent EKOS poll found "the incidence of [Canadians] thinking there are too many visible minorities is up significantly and no longer trails opposition to general immigration (as it has historically)."

Here is a picture that shows the question EKOS asked, and the long-term trend in Canadians' responses to this question (click on the picture to make it bigger):

The poll generated a certain amount of discussion (e.g. here and here). Andray Domise, writing in Macleans, observed:

And given that Canada‘s immigrant population is fairly broadly distributed between a range of countries (unlike the U.S., for example, where the plurality of immigrants come from Mexico), there is no real basis in fact for this “too many visible minorities” sentiment. Only the deeply wonkish would have a working knowledge of the ethnic makeup of Canada’s immigrant population, which effectively made EKOS’ poll a litmus test on Canadian racism.

Deeply wonkish? We're here, and we hear your call!

Do Chinese Canadians struggle in the labour market?

Update: A revised version of the paper discussed in this blog, with results close to those obtained using the PUMF, is available here. Thanks to Feng Hou for taking my concerns seriously and responding to them promptly.