Skip navigation
Margin Notes

The problem with PhD training, take 2

But PhDs grads are getting the jobs and earning more, aren’t they?


My most recent blog post and accompanying news story on the problem with PhD training in Canada got many views and received many thoughtful comments. It appears to have touched a nerve. Most who commented seemed to agree in principal with the general premise that the content of PhD programs needs to be expanded beyond training these students for jobs in academia, while others argued that we are simply producing too many PhDs.

However, on a somewhat different note, a couple of tweets from Joey Berger questioned the concerns that PhD students and postdocs have about finding jobs. Mr. Berger linked specifically to two charts from the National Graduates Survey of Statistics Canada.

The first chart shows that individuals who graduated with a doctoral degree in 2005 and who were working full-time two years later were doing quite well in terms of earnings, with the median salary ranging between $60,000 and $85,000, depending on the field of study.

I tweeted him back, saying that’s fine if you’re working full time. To that, Mr. Berger responded with a link to a second chart from the same survey showing rates of full-time employment by level of study. “Why does the conversation about PhD outcomes almost never include this data?” he asked.

I can’t say that I find the second chart makes a compelling case. At a glance, the rate of full-time employment for those with a doctoral degree in certain fields does not appear to be any better than those with an undergraduate or master’s degree.

Furthermore, there is another chart from that same survey which is even more interesting. This chart shows earnings distribution of 2005 graduates working full-time in 2007 by gender and level of study. Leaving the gender aspect aside, the chart clearly shows that the greatest earnings premium is between the bachelor’s and master’s level, and that the premium from the master’s to doctoral level is low to non-existent. Here is how StatsCan characterizes it:

The largest earnings premium existed between the bachelor and master levels suggesting that investing in further post-graduate work is financially beneficial. On the other hand, the earnings premium between a master level and doctorate level suggests that the monetary gain from employment two years after graduation for doctorate students is marginal.

These are not new data and have been hashed out before. But, they do appear to back up concerns about the value of a PhD as it currently exists. Of course, there are lots of caveats. The earnings premiums and employment levels may look very different 10 years out compared to just two years after graduation. What’s more, the economic situation now is quite a bit more precarious than it was in 2007, which could affect the relative standing of PhD grads.

As always, your thoughts, rebuttals or other comments are greatly appreciated.

Léo Charbonneau
Léo Charbonneau is the editor of University Affairs.
Post a comment
University Affairs moderates all comments according to the following guidelines. If approved, comments generally appear within one business day. We may republish particularly insightful remarks in our print edition or elsewhere.

Your email address will not be published. Required fields are marked *

  1. dr.doinglittle / December 8, 2011 at 15:54

    According to the second chart, a person with an MA in the humanities actually has a greater chance of getting a FT job than someone with a PhD. And in either case, it’s below 70 percent.

    So, 6-15 years of education and paying tuition so you can have slightly better than a 2/3 chance of getting a job, and you’ll be well into your 30s…. Brutal.

  2. Joey Berger / December 9, 2011 at 14:48

    Hi Léo,

    Thanks for following up on this. I think your commentary underscores my point, which is that discussions about graduate outcomes should be informed by data, especially when that data is robust, relevant and free – as is the case with the NGS figures in the two charts linked to above. The point of my tweet(s), it’s worth keeping in mind, wasn’t to argue that PhDs are fine – it was to question whether a panel discussion on graduate outcomes at a science policy conference made was informed by this readily available data. The fact that it wasn’t doesn’t mean that the conclusions about the precarious prospects of humanities grads aren’t legitimate (it’s not a debating contest, after all). But it does mean that the discussion is ill-served.

    Kudos to you for engaging in the figures!


  3. David Kent / December 10, 2011 at 10:05

    Hi Joey,

    David Kent here (I ran the panel at the CSPC that Léo has been referring to). In their presentations, the panelists were very good about providing statistical evidence (though the NGS results were not specifically raised).

    Angela Crawley from the Canadian Association of Postdoctoral Scholars (CAPS) presented key portions of the data from the CAPS 2009 survey of ~20% of total Canadian postdocs and additional data from studies in the United States concerning the success rates of postdocs and expectations of postdocs (which I’ve just written about on my own site at The vast majority of this info can be found on the CAPS website.

    In addition to Angela’s presentation, Olga Stachova from MITACS presented OECD statistics from the Programme for International Student Assessment.

    I hope this helps a little Joey – I’m also happy to provide references for any of the info that we may present on the site or at meetings.

  4. Gail Matthews / December 12, 2011 at 16:18

    Hi Léo,
    I enrolled in the PhD after doing my research and learning there was a need in the field of education and that I could support teachers and students. In 2004 I earned a PhD in Education and I left my home province for a year of full-time university contract work. After returning I spent 15 months of searching for all sorts of jobs, especially entry-level as I was advised it was the way to “get your foot in the door” e.g., reception, sales clerk, administration. Later, I found a 3 year contract job. However, I am unemployed again and have been job searching for the past 14 months. I do a lot of volunteer work. In the daytime I’ve volunteered in schools (I am a certified member of the Ontario College of Teachers) and in community organizations in the evenings. I found a 16-week job contract and put in additional hours to make certain the work was of an excellent calibre. Needless to say, a few people were extremely abusive once they learned about the PhD (not from me). Not complaining. I am sharing one experience of the reality of work in the “real world”. Education does NOT entitle anyone to an income or a job with pay. It is expected that one do free work in the real world, and as a highly educated person with a PhD with a natural inclination and commitment to make the world a better place, I do free work (the fancy word is volunteer) because I have superior skills, knowledge, an excellent background and an excellent work ethic. I do free work because it benefits organizations and people (like professors) who will never be able to afford it. By the way, well-meaning professors have benefitted from my free work while I was looking for paid work and unable to feed myself. Not complaining, I am sharing one experience with other people out there with PhDs looking for work. There is a notion that “who you know” is the way jobs are found. In the university there are a lot of well meaning people who give a lot of free advice, all talk and no action. Your free work is just that free work with a fancy name called volunteering. Free work does not lead to a toe in the door. Free work may lead to a job with entry level pay and university colleagues who will make certain that you know that they are doing you a favour. Through all of these events I am expected to and continue to repay my debt and mortgage. In the meantime, I am on welfare and use food banks because I can no longer afford to feed myself. Happy Christmas to everyone!

  5. S / December 13, 2011 at 12:56

    Not sure about the humanities, but in computer sci, biomedical and economics- you need a PhD to be taken seriously. If the FT work rates are similar and PhD holders with FT positions are doing pretty well, then things must be better than they are made out to be.