One enormous myth that has been propagated (sometimes innocently, sometimes not) in recent debates over the future of the auto industry is the false notion that auto workers “make” $75 per hour.
Autoworkers don’t remotely make that much money — yet the lie has been repeated often enough, I am amazed at how many people actually seem to believe it. In reality, CAW-represented production workers in the major auto plants are paid a starting wage of about $24 per hour. This grows (as they earn seniority) to a maximum of $34 per hour. Skilled-trades specialists (in electrical, machining, tool, or related trades) earn up to $40 per hour. On a weighted average basis (counting production and trades together), the average is $35.
Workers at non-union auto plants in Canada earn almost exactly the same. And that will remain true, so long as management at those companies is sufficiently concerned about unionization that they pay a union wage to their non-union workers — or better yet, until they decide to join the union that indirectly accounts for their wages. (This also proves why it’s self-defeating to try to compete with non-union factories by cutting wages: obviously Toyota and Honda will also “follow the pattern,” and cut their own wages. Honda Canada already did this, announcing this month a wage freeze for production workers to match the equivalent freeze negotiated by the CAW with GM and Chrysler.)
A workers’ compensation, of course, includes their non-wage pension and benefits. For CAW members, the total annual cost of their pension plan (current service), health benefits (including the current service expense for their post-retirement health benefits), and other benefits (tuition assistance, car purchase incentives, etc.) adds up to less than $20,000 per worker per year. That works out to under $9 per hour (for a full-time worker employed 40 hours per week, 52 weeks per year).
An autoworker’s true “compensation,” therefore, is around $44 per hour ($35 in wage, and $9 in benefits). That’s more than most workers, obviously. But less than many in our society (lawyers? stock brokers? political aides?). And not remotely equal to the inflated numbers that have been cited so widely. Autoworkers’ jobs are challenging (both physically and mentally), and highly productive (GDP per person in the auto assembply sector is $300,000 per year). CAW compensation is also internationally competitive, comapred to other developed automaking jurisdictions (and lower than averages in Germany, the U.S., and Japan).
So where does the idea come from that auto workers make over $70 per hour?
That claim refers to a concept that is very different from someone’s hourly wage. It is not even equal to someone’s total compensation. It refers to a measure called “all-in hourly labour cost.”
All-in hourly cost is calculated by dividing all labour-related expenses by the number of hours actually worked in a year. It thus includes all kinds of things that wouldn’t normally be considered part of someone’s hourly compensation. This includes:
- The pro-rated cost of paid time off (which reduces the denominator of the equation, by reducing hours worked).
- The effective cost of other absences from work (CAW members don’t get paid when they are sick, until their S&A insurance eventually kicks in — but illness still increases the hourly cost, by virtue of reducing the number of hours worked).
- The effective cost of downtime and layoffs (which also reduce hours worked). This one is especially infuriating: not only do CAW members experience the loss of unemployment, it also “looks” like they’ve become more “expensive” as a result!
- The cost of payroll taxes paid by the employer to government (CPP, EI, health tax, and workers’ comp premiums). Is this compensation? Hardly. And how ironic that the Boss’s taxes are counted as part of the Worker’s income!!
- The per-hour expense associated with changes in pension funding status and changes in the actuarial assumptions used to estimate the cost of retiree health benefits. These factors (the ups and downs of the stock market, for example) cause enormous swings in the cost of these so-called “legacy costs” from one year to the next. And at any rate, these legacy costs have nothing to do with the people actually working in factories today.
For all of these reasons, the all-in labour cost measure that’s been so widely cited in the paper is an incredibly misleading portrayal of autoworkers’ actual compensation. But this misstatement has played its purpose, motivating some of the hateful phone-in radio shows and on-line blogs that have added to the considerable political heat on the CAW’s recent bargaining.
A full deconstruction of the all-in labour cost methodology is provided in this recent CAW research paper:
Just for fun, try calculating the all-in labour cost for someone you know. I tried it on my spouse, who is a university professor. I added her healthy salary to her current service pension cost and other benefits and payroll taxes. I threw in an extra amount to reflect the pension deficit that her university is now grappling with (thanks to the market meltdown). I divided that by the number of hours she teaches in class per year, prepares her lectures, or sits in stupid committee meetings. [I understand this is parody of the hard, long work that committed academics actually perform, but bear with me.] Then I adjusted those hours to reflect every seventh year being a sabbattical.
My answer: over $600 per hour all-in labour cost. What a joke. No-one would believe this is what professors actually get “paid.” So why do they believe it of autoworkers?