The Rate of Exploitation of General Motors Workers

I thought it would be politically relevant to try to estimate the rate of exploitation of General Motors (GM) workers for 2019 (since annual reports starting in 2020 would distort the picture because of the pandemic). I say politically relevant because of the closure of the GM Oshawa plant on December 18, 2019 and the subsequent making of the document Company Town, which dealt with the coming closure, the attitude of Jerry Dias, president of Unifor (the union that represents the workers at Oshawa) and the consequences of the closing of the factory.

However, GM annual reports (like many annual reports based in the United States), provide insufficient information to calculate the rate of exploitation. For example, there are no data on wages and salaries paid out (although there are for benefits).

Nonetheless, I searched for substitutes for the data. Undoubtedly, such calculations will be even more imperfect than the rates of exploitation I calculated for various large employers in Canada. It will undoubtedly only include bare statistics, without much refinement and with few adjustments. Still, such estimates may provide a ballpark figure of the extent of exploitation.

I invite others to criticize the data used and the manner of determining the rate of exploitation–by providing more accurate data and a more accurate manner of determining the rate of exploitation.

Where possible, I provide the website addresses where I found the information if the information is not drawn from the Annual Report.

The Nature of the Rate of Exploitation

But what is the rate of exploitation? And why not use the usual rate of profit or the rate of return? The rate of profit is calculated as profit divided by investment. Since employers purchase both the means for work–buildings, computers, office supplies, raw material–and hire workers–we can classify investment into two categories: c, meaning constant capital, or the capital invested in commodities other than workers; and v, or variable capital, the capital invested in the hiring of workers for a certain period of time (wages, salaries and benefits).

The purpose of investment in a capitalist economy is to obtain more money (see The Money Circuit of Capital), and the additional money is surplus value when it is related to its source: workers working for more time than what they cost to produce themselves. The relation between surplus value and variable capital (or wages and salaries) is the rate of surplus value or the rate of exploitation, expressed as a ratio: s/v.

When the surplus is related to both c and v and expressed as a ratio, it is the rate of profit: s/(c+v).

In Marxian economics, you cannot simply use the economic classifications provided by employers and governments since such classifications often hide the nature of the social world in which we live. The rate of profit underestimates the rate of exploitation since the surplus value is related to total investment and not just to the workers. Furthermore, it makes the surplus value appear to derive from both constant capital and variable capital.

I decided to look at the annual report of some of the largest private companies (if they are available) in order to calculate the rate of exploitation at a more micro level than aggregate rates of surplus value at the national or international level. Politically, this is necessary since social democrats here in Toronto (and undoubtedly elsewhere) vaguely may refer to exploitation–while simultaneously and contradictorily referring to “decent work” and “fair contracts.” Calculating even approximately the rate of exploitation at a more micro level thus has political relevance.

Conclusions First

As usual, I start with the conclusion in order to make readily accessible the results of the calculations for those who are more interested in the results than in how to obtain them.

We now have sufficient information to calculate the rate of exploitation of GM workers.

Adjusted Income before income taxes: $7.383 billion=s
Total wages and benefits $18.597 billion=v

To calculate the rate of surplus value or the rate of exploitaiton (they are the same thing), we need to divide “Adjusted Income before income taxes” (s) by “Total wages and benefits” (v).

So, with the adjustments in place, the rate of exploitation or the rate of surplus value=s/v=7.383/18.5976=40%.

That means that for every hour worked that produces her/his wage, a worker at GM works around an additional 24 minutes for free for GM.

In a 7-hour (420-minute) work day , the GM worker produces her/his wage in about 300 (5 hours) and works 120 minutes (2 hours) for free for GM. Of course, during the time that the worker produces her/his own wage, s/he is subject to the power of management and hence is unfree (see, for instance, Management Rights, Part Four: Private Sector Collective Agreement, Ontario and Employers as Dictators, Part One).

In an 8-hour (480 minute-work day), a GM worker produces her/his wage in 343 minutes (5 hours 43 minutes) and works for 137 minutes (2 hours 17 minutes) free for GM.

In an 9-hour (540-minute) day, a GM worker produces her/his wage in 386 minutes (6 hours 26 minutes) and works for free for 154 minutes (2 hours 34 minutes) for GM.

In a 10-hour (600-minute) day, a GM worker produces her/his wage in 429 minutes (7 hours 9 minutes and works for free for 171 minutes (2 hours 51 minutes) for GM.

In a 11-hour (660-minute) day, a GM worker produces her/his wage in 471 minutes (7 hours 51 minutes) and works for free for 189 minutes (3 hours 9 minutes) for GM.

In a 12-hour (720-minute) day, a GM worker produces her/his wage in 514 minutes (8 hours 34 minutes) and works for free for 206 minutes (3 hours 26 minutes) for GM.

Of course, during these times that the worker works to obtain an equivalent of her/his own wage, s/he is subject to the power of management and hence is also unfree during that time (see The Rate of Exploitation of Magna International Inc., One of the Largest Private Employers in Toronto, Part Two, Or: Intensified Oppression and Exploitation and Employers as Dictators, Part One).

Many GM workers in the United States (and in Canada) belong to a union. The Annual Report states:

At December 31, 2019 approximately 48,000 (50%) of our U.S. employees were represented by unions, a majority of which were represented by the International Union, United Automobile, Aerospace and Agriculture Implement Workers of America (UAW).

Despite belonging to a union, the GM workers are exploited–but to a relatively low extent–much lower than many other union workers. The highest calculated rate of exploitation so far has been Rogers Communications’ workers, at 209 percent (see the comparative rates in the post The Rate of Exploitation of Workers at WestJet Airlines Ltd.). Why that is would be a good area for research.

Political Questions

Do you think that these facts contradict the talk by the left and unionists of “fair wages,” “fair contracts” (see Fair Contracts or Collective Agreements: The Ideological Rhetoric of Canadian Unions, Part Three: Unifor (Largest Private Union in Canada) for the rhetoric of the largest Unifor as the largest private-sector union in Canada, and “decent work?” Do they ignore the reality of life for workers, whether unionized or non-unionized?

What of the following, drawn from the November 5, 2020 collective agrement between General Motors of Canada Company and Unifor Local No 199 St. Catharines, Local No. 222 Oshawa and Local No. 636 Woodstock? Page 7:

Section IV

Management

(4) The Union recognizes the right of the Company to hire, promote, transfer, demote and lay off and to suspend, discharge or otherwise discipline employees for just cause subject to the right of any employee to lodge a grievance in the manner and to the extent as herein provided.

The Union further recognizes the right of the Company to operate and manage its business in all respects, to maintain order and efficiency in its plants, and to determine the location of its plants, the products to be manufactured, the scheduling of its production and its methods, processes, and means of manufacturing. The Union further acknowledges that the Company has the right to make and alter, from time to time, rules and regulations to be observed by employees, which rules and regulations shall not be inconsistent with the provisions of this Agreement.

This power of management is not mentioned at all in the National Unifor Bargaining Report. Like most union bargaining reports, it omits all negative aspects of working for General Motors (including being exploited and oppressed):

HIGHLIGHTS
• $1.1B to $1.4B in investments
• General wage increases
• $7,250 Productivity and Quality Bonus
• Inflation Protection Bonuses
• Improved New Hire Program
• Skilled Trades Adjustment
• Benefit improvements
• Three-year term
• Lump sum payment for pre-1994 retirees

A written summary also omits the continued power of GM management to exploit and oppress workers (page 1):

JOINT MESSAGE TO ALL GENERAL MOTORS MEMBERS

SECURING A MADE IN CANADA FUTURE

If there is a lesson learned from 2020 Auto Talks, it is that the future of Canada’s auto sector is bright and on a clear forward path.

Thanks to the hard work and determination of the Unifor-GM Master Bargaining Committee, we are proud to present a new collective agreement that follows the economic pattern negotiated at both Ford and FCA. This agreement includes a 5 per cent increase to hourly wages, a 4 per cent lump sum payment in 2021, along with $11,250 in bonuses.

The deal makes major improvements to the New Hire Program, including an accelerated path to full rate, and returns key benefits like the Legal Services Plan and the afternoon (5%) and midnight (10%) shift premium.

Skilled trades workers will see their 20% wage differential restored, new apprentices hired, and the pre-apprenticeship program re-instated for future hires. The new agreement also includes significant improvements to the benefits plan, modest (but still important) pension improvements, along with health and safety gains, retirement allowances and equity gains including 10 days of paid domestic violence leave and a new Racial Justice Advocate.

Along with these contractual improvements are commitments by the company to maintain and expand work at current Unifor facilities. GM will continue V6 engine and 6-speed transmission production over the life of the contract, and forecasted beyond.

Also, in a stand-alone letter GM has committed to explore new potential product programs and investment opportunities for St. Catharines, with input from them union. [my emphasis. When I formed part of a negotiating team for Operating Engineers Local 858, in Prince George, British Columbia, the management team were able to shuffle off many items on the negotiating table by referring to a “consultation process” between the union and management–in effect eliminating such items for negotiations. The verb “explore” and the noun “input” are euphemisms for the right of management to simply do what it wants, with the proviso that it “consults” the union.]

St. Catharines is well regarded as a leader in the GM powertrain division and will receive $109 million to in-source new transmission work for the Corvette, adding jobs, and make
upgrades to the small block V8 engine program. GM will continue V6 engine and 6-speed transmission production over the life of the contract, and further commits to seek out new
programs that sustain the facility over the long term.

The Woodstock PDC will receive $500,000 in additional upgrades. Aftermarket parts work at Oshawa will also continue, maintaining hundreds of jobs.

In addition, and pending ratification, GM has committed to invest up to $1.3 billion to restart pickup truck assembly at the Oshawa Assembly Complex, with an expected two-shift operation in the first half of 2022 (and the potential for a third).

These “highlights” teach the workers nothing about the limitations of collective bargaining and collective agreements. They are designed to hide the concentration of major decision-making power in the hands of General Motors (such as the “right of the Company to determine the location of its plants” and the lack of such power by unionized workers.

The same could be said of the Local 222 Bargaining Report, which recommended voting for the collective agreement without any explicit indications of its limitation as indicated in the management rights clause of the collective agreement. Thus, the Report indicates among other things, the following (page 3):

Commitment to settling the 2020 GM/Unifor Master Agreement and Oshawa Local Agreement
•The production allocation is for the current life cycle. Currently, there is no future product commitment but the Company has expressed that the life cycle will be a minimum of three (3) years and that is well into the new Collective Agreement 2023.

•There will be no retirement incentives offered at the Oshawa Assembly Plant during the current life cycle of the product. In the event of a permanent reduction in force, the new hires at the Oshawa Assembly Plant will be laid off. Any employees hired prior to the 2020 Collective Agreement will flow back into the Oshawa OEM Stamped Products and Service Operation based upon Seniority.

Of course, workers have to subordinate their will to the will of employers in a society dominated by a class of employers, and so no union representatives can overcome this limitation; such a limitation is a class limitation, and it is at this level that such limitations need to be addressed. However, the class level is hardly a level that excludes the particular sections of the working class. Those particular sections are included in that general level, so at the local, regional or national level, the class issue can certainly be indicated and not simply ignored–which is what union reps do often enough these days. At the least, they could explicitly indicate the limitations of the collective-bargaining process and the collective agreements that result from that process. Better yet, they could not only include such limitations, but they could point to ways in which such limitations might be overcome through regional, national and international tactics and stragegies. Most modern union reps, however, have no intention of doing so; indeed, they are likely unaware of the need to do so.

Data on Which the Calculation Is Based

The calculation of the rate of exploitation is undoubtedly imperfect, and I invite the reader to correct its gaps. Nonetheless, the lack of any attempt to determine the rate of exploitation at the city level has undoubtedly reinforced social-reformist tendencies. (in millions of U.S. dollars)

Total net sales and revenue 137,237

Total costs and expenses 131,756

Operating income 5,481 [137,237-131,756=5481]

Adjustments of Surplus Value (Profit)

It is necessary to make some adjustments to this since the annual report also refers to the following additional categories:

Automotive interest expense 782
Interest income and other non-operating income, net 1,469
Equity income (Note 8) 1,268

Starting with the category “Automotive interest expense,” it is necessary to make an adjustment. It is necessary to add 782 (Automotive interest expense) to “Operating income” since (as I explained in another post):

In Marxian theory, it is necessary to question whether some expenses are expenses for both the individual employer and for the class of employers (and fractions of their class, such as those who live on interest); in such a case, the expense is deducted from total revenue. On the other hand, there are expenses that are expenses for the individual employer but are not expenses when looked at from the point of view of the class of employers; in such an instance, they are paid out from the surplus value produced or obtained by workers and are to be included in income before taxes.

Accordingly:

Temporarily adjusted income before taxes 5,481+782=6,263

Moving to the next category, “Interest income and other non-operating income, net,” since the idea of calculating the rate of exploitation of particular employers is to determine the extent to which the particular employer exploits its workers, income derived from the exploitation of workers other than its workers should be excluded. I did not think this through or consider this when I calculated the rate of exploitation of some other employers (such as Air Canada); I may or may not recalculate the rate of exploitation of such employers in the future–that depends on how much time I have to dedicate to writing other posts and engaging in my own research as well as my own personal commitments to my daughter and wife.

In the particular case of General Motors, I will exclude such income from the calculation since the income is derived from workers other than the workers of GM.

The last category, “Equity income,” seems to reflect net surplus value after expenses are subtracted from revenue. Note 8 elaborates:

Note 8. Equity in Net Assets of Nonconsolidated Affiliates

Nonconsolidated affiliates are entities in which we maintain an equity ownership interest and for which we use the equity method of accounting due to our ability to exert significant influence over decisions relating to their operating and fmancial affairs. Revenue and expenses of our joint ventures are not consolidated into our financial statements; rather, our proportionate share of the earnings of each joint venture is reflected as Equity income.

Since “Equity income” reflects the “proportionate share of the earnings of each joint venture,” it constitutes the net result of GM exploiting workers in joint ventures. Accordingly, it is necessary to make an adjustment. It is necessary to add 1,268 to Income before income taxes. The final adjustment is:

Adjusted income before taxes 7,531

Wages and Salaries (v)

Although there are statistics in the annual report for employee benefits, there are no statistics in it for wages or salaries.

I failed to find any direct information of total salaries, wages and benefits on the Net. The best that I could do was to find data about the total number of employees and then try to find data on the average wage/salary as well as average benefits and multiply the sum of the average wage/salary and benefits by the total number of employees.

To improve such calculations, I invite Sam Gindin, former research director to the Canadian Auto Workers (now Unifor) or Jim Stanford, former economist for the same union, to provide more accurate data (perhaps insider data?).

In 2019, there were 164,000 employees:

Employees At December 31, 2019 we employed approximately 95,000 (58%) hourly employees and approximately 69,000(42%) salaried employees.

This is consistent with the following:

( https://www.macrotrends.net/stocks/charts/GM/general-motors/number-of-employees#:~:text=General%20Motors%20total%20number%20of%20employees%20in%202020%20was%20155%2C000,a%203.89%25%20decline%20from%202017).

This number, as I argued above, needs to be multiplied by the average cost of a GM worker, including benefits. According to some, the average cost to use a GM worker in 2019 was about $63 US an hour. From https://www.spglobal.com/marketintelligence/en/news-insights/trending/Fic7Dwvvxuh14hs9rmjwJw2, dated January 15, 2020:

GM’s average hourly labor costs are estimated to be … $63 in 2019.

I assume that this includes benefits. This amount is less than the amount estimated in 2006 (or perhaps 2009–it is unclear). From

From https://bigthreeauto.procon.org/view.additional-resource.php?resourceID=002050 :

Average Hourly Compensation 2006 (US Wages and Benefits)
Last updated on: 1/21/2009 6:47:00 AM PST“The calculations show, accurately enough, that for every hour a unionized worker puts in, one of the Big Three really does spend about $73 on compensation. So the number isn’t made up. But it is the combination of three very different categories.

The first category is simply cash payments, which is what many people imagine when they hear the word ‘compensation.’ It includes wages, overtime and vacation pay, and comes to about $40 an hour. (The numbers vary a bit by company and year. That’s why $73 is sometimes $70 or $77.)

The second category is fringe benefits, like health insurance and pensions. These benefits have real value, even if they don’t show up on a weekly paycheck. At the Big Three, the benefits amount to $15 an hour or so.

Add the two together, and you get the true hourly compensation of Detroit’s unionized work force: roughly $55 an hour. It’s a little more than twice as much as the typical American worker makes, benefits included. The more relevant comparison, though, is probably to Honda;s or Toyota’s (nonunionized) workers. They make in the neighborhood of $45 an hour, and most of the gap stems from their less generous benefits.

The third category is the cost of benefits for retirees. These are essentially fixed costs that have no relation to how many vehicles the companies make. But they are a real cost, so the companies add them into the mix — dividing those costs by the total hours of the current work force, to get a figure of $15 or so — and end up at roughly $70 an hour.”

As to be expected, the estimated $63 an hour ($US) is certainly much higher than the estimated hourly wage of American auto workers. From Automotive Industry Labour Market Analysis: Wage Report (Canadian Skills Training and Employment Coalition, Prism Economics and Analysis, and the Automotive Policy Research Centre, October 2019), page 30:

Research done by the Center for Automotive Research (CAR) found that nominal wages for U.S. workers in the auto manufacturing sector increased by just over 6% from $28.49 in 2002 to $30.20 in 2018 but real wages fell by 23.5% (as cited in Haglund, 2019).

The equivalent of $63 per hour cost of GM workers is consistent with the following (From cached https://webcache.googleusercontent.com/search?q=cache:mQ2Z0vRWOxQJ:https://abcnews.go.com/Business/wireStory/tank-uaw-deals-increased-automakers-labor-costs-68309323+&cd=1&hl=en&ct=clnk&gl=ca):

Think tank says UAW deals increased automakers’ labor costs

New contracts between the United Auto Workers union and Detroit’s three automakers substantially increased the cost gap between Detroit and foreign automakers with U.S. factories

ByThe Associated Press

January 15, 2020, 3:40 PM

…

Total labor costs include wages, health care, pensions and other expenses.

Center Vice President Kristin Dziczek calculated that GM’s labor costs would rise from $63 per hour before the new contract to $71.

It is also consistent with the following (https://www.spglobal.com/marketintelligence/en/news-insights/trending/Fic7Dwvvxuh14hs9rmjwJw2#:~:text=GM’s%20average%20hourly%20labor%20costs,21.1%25%20from%20%2457%20in%202015.):

GM’s average hourly labor costs are estimated to be $71 in 2023, up 12.7% from $63 in 2019 and up 29.1% from $55 in 2015, Dziczek said during a Jan. 15 webcast. Ford’s hourly labor costs would be $69 in 2023, up 13.1% from $61 in 2019 and up 21.1% from $57 in 2015. Fiat Chrysler’s costs would go up to $66 in 2023, up 20% from $55 in 2019 and up 40.4% from $47 in 2015.

Given the above, $63 an hour seems to be the average cost of a GM employee whereever s/he works. There will be no adjustments for this category despite the above calculated reduction of 173.5 from “Tranformation activities” due to some of that category involving separation benefits received by workers since, presumably, the $63 an hour includes such separation benefits.

Jim Stanford’s Disagreement with the Estimate of $63 an Hour Cost of Variable Capital V)

However, Jim Stanford would dispute such an hourly cost since he disputes the hourly cost of $75 (see How Much do Autoworkers REALLY Make? Surprise: It’s NOT $75 Per Hour! http://unifor584retirees.ca/caw_retirees/pdf/hourly_labour_costs_09.pdf). He argues that workers receive between $43 and $44 hour.

Before I delve into this issue, let me preface it with the purpose of calculating the rate of exploitation. It is supposed to determine the proportion of hours worked that produce the value of the wage received by the workers in relation to the hours worked for no compensation and thus free for the employer. The wage is a composite of the actual wage rate and benefits and cannot be limited to the given wage rate.

Generally, the rate of exploitation is a class phenomenon, and the value of labour power or variable capital (v) and surplus value as its components are also class phenomenon. As Ben Fine states, in relation to the value of labour power, pages 104-105:

… for Marx the value of labour power is the consequence of an exchange between capital and labour, confronting each other as the two major economic classes. It is not simply the wage earned on the labour-market by one individual as opposed to another. …

the value of labour power is a more complex concept than the wage rate or earnings
of the typical worker.

Some of the costs of variable capital for employers are class costs, or the costs to the class of employers and apply across the board to all employers (or at least to a section of employers in a particular industry). Workers may not individually receive them, but some workers do; which workers depends on various conditions, such as the level of unemployment, the age of the workers, their health and so on. They need to be included in the value of labour power or variable capital even if no specific set of workers receive them as a benefit since they are costs for the employer being an employer of any set of workers whatsoever.

Stanford, however, excludes several categories which is included in the above $63 an hour–a category called “All-in Active Hourly Labour Cost.” The subcategories of this category and the corresponding amounts are:

Overtime and shift premiums $3
Cost per hour worked of paid time off $8-9
Impact on hourly cost of layoffs & downtime $1-3
Cost per hour worked of SUB $1-3
Statutory taxes $3-4

Overtime and shfit premiums

Stanford excludes this subcategory from his calculation on the basis of the following:

Occasionally companies will require their workers to stay overtime, beyond normal working hours. Overtime is worked ins response to surges in consumer demand, to make up for production problems or bottlenecks, or in some cases because employers have decided it’s cheaper to work its existing staff extra hours than to hire new workers. In every case, it is the employer’s choice when overtime is worked.

Workers required to work overtime are paid a wage premium (usually 50%) for those hours. To a large extent (depending on the specific hours involved), overtime pay is mandated by labour law (although a labour contract can require overtime to be paid in some circumstances when it is not legally required).

In addition, in the auto industry and other manufacturing settings, it is standard practice to pay a shift premium for workers who staff evening and night shifts. (In CAW-represented auto plants, there is a 5% premium paid for evening shifts, and a 10% premium for overnight shifts, to reflect the added stress on family life of those working hours.)

Is the overtime premium part of one’s “hourly wage”? Few Canadians would conceive it that way – although those working overtime certainly appreciate the extra money. And remember: overtime is

something that occurs because employers desire it.

According to Statistics Canada, in 2007 (most recent data), auto assembly workers in Canada worked an average of about 3.5 hours of overtime per week. This increased total average wage payments (weighted across all hours worked in the year) by around $2. Shift premiums added, on average, about another dollar per hour.

This argument is unconvincing. Overtime is supposed to not constitute compensation, he implies, because it is not voluntary. Being voluntary or involuntary has nothing directly to do with the cost of workers. How workers conceive overtime also hardly determines whether it is a cost. As for shift premiums, the same logic applies; they too are costs. Stanford never indicates what overtime payments and shift premiums objectively are.

We can, however, get some idea of what they are by referring to the value of labour power as subject to a normal working day under average conditions. Since overtime work in effect extends the working day beyond the norm, it involves abnormal consumption of the labour power of workers. As Marc Linder (2000) argues ( “Moments are the elements of profit”: overtime and the deregulation of working hours under the Fair Labor Standards Act, pages 5-6):

Marx also furnished a general framework for understanding struggles over the length of the workday or workweek. On the surface, this struggle centers on the conflict between the buyer and the seller of a commodity which generates special problems because, unlike the situation with a general run-of-the-mill commodity, the body and mind of the human seller of labor power cannot be separated from its daily use by the buyer. Since the law of exchange of commodities, however, does not recognize any special rules for this particular exchange, the capitalist buyer tries to extract the greatest possible profit from the use of the worker’s labor power for the day’s or week’s worth he has bought. The question then becomes: how long is a workday or workweek? Since the human seller lives beyond the day, he must make sure that he sells his only commodity for a price high enough to enable him to reappear at work the next day with his labor power in a condition of strength and health that meets the standards set by his competitors. But the worker as a rational labor market participant must also exercise sufficient foresight to husband his only economic asset for a lifetime—or at least the standard working life of his type of labor. If the daily value of his commodity equals its lifetime value divided by 30 years or approximately 10,000 workdays, then he must make sure that overlong workdays and workweeks do not force him to expend so much additional energy that he uses up 1/5,000 or 1/3,333
of his lifetime supply for only 1/10,000 of its lifetime value. For this reason socialist unions regarded eight-hour laws as “life lengthening” acts.The worker therefore regards such overwork as crossing the line from the capitalist’s rightful use to plundering of his labor power and, as such, a breach of their contract and of the law of the exchange of commodities. His demand for a workday or workweek of normal length—defined by its compatibility with a healthy 30-year worklife—is as rightful as the capitalist’s demand that the worker work as long as possible each day and week. Because the capitalist is not a slaveholder, he has no (capital-) invested interest in the length of the worklife of his individual employees: “A quick succession of unhealthy and short-lived generations will keep the labour market as well supplied as a series of vigorous and long-lived generations.” Thus as long as the employer can find equivalent replacements in the labor market when he needs them, this private contractual dispute cannot be resolved between individual buyer and seller. The resulting “antinomy” of right against right27 must, Marx argued, be decided by “the respective powers of the combatants.” But since “in its merely economic action capital is the stronger side,” a class-wide settlement of the hours issue was possible only through “general political action,” which meant “legislative interference” under pressure from the working class.2* Consequently, the normalization of the workday and workweek appears historically as a struggle between the “aggregate capitalist, i.e., the class of capitalists, and the aggregate worker, or the working class.”

Overtime and shift work can be conceived as compensatory costs for abnormal consumption of workers–but they are costs of hiring workers, exploiting them and oppressing them.

Cost per hour worked of paid time off

Stanford has the following to say on the matter in order to justify excluding this subcategory:

Now here is where it starts to get more complicated. “All-in hourly labour cost,” in the auto industry, is not calculated by dividing total compensation by the number of normal working hours in a year (as we have done above: 40 hours per week
times 52 weeks in a year equals 2080 working hours in a year).

Let us pause here. What Stanford calls “normal working hours” is not the statistic that can be used to calculate the rate of exploitation since it is not the number of actual hours worked. It is the actual hours worked that produces the equivalent of the total compensation received by workers that is relevant–and not some “normal working hours” that no workers actually work.

From Anwar Shaikh and E. Ahmet Tonak (1994), From Measuring the wealth of nations: The political economy of national accounts, page 178:

By definition, Marxian labor value added is simply the number of hours worked by productive workers

Let us continue with Stanford’s views.

Instead, all-in hourly labour costs are calculated over a much smaller base of hours. Total
compensation costs are divided by the numbers of hours actually worked in a year. Actual hours worked, the denominator of this fraction, differs from the number of standard working hours in a year (2080) for several reasons:

Paid time off (for vacation and holidays)

Sick leave (CAW autoworkers do not receive any pay during the first days of
an illness, after which they are compensated under a sickness and accident
insurance scheme)

Time not worked due to layoffs or downtimeIt is simple mathematics that the lower is the number of hours actually worked,
the higher is the apparent “all-in hourly labour cost.”

The CAW has placed great emphasis over the years on negotiating more paid time off, as a deliberate strategy to try to protect employment levels against the effects of technological change and productivity growth, and to provide for needed time away from the physical and mental stresses of assembly line work. However, in recent contracts the amount of paid time off has been reduced by 80 hours per year (in the face of intense cost-cutting pressure from the employers). Today a CAW production worker with maximum seniority (over 20 years) qualifies for 6 weeks of paid time off (for vacation, scheduled mandatory vacation or “SPA,” and personal leave). A new hire qualifies for 2 weeks (the legal minimum). A worker with 5 or more years seniority qualifies for 4 weeks.

Holidays (including regular statutory holidays and a week-long Christmas shutdown) reduce working time by another 15 days per year.

Paid time off can be considered a form of compensation. It can also be considered a basic human and labour right – one that workers have fought for over the centuries, and that is essential to the quality of life of working people and their families. Of the paid time off received by CAW autoworkers, about half is required by law. The rest reflects additional time negotiated by the union. I doubt, however, that many Canadians consider their paid time off as part of their hourly wage. They conceptualize it separately, as time. Someone who earns $15 per hour, but is allowed to take a total of five weeks off per year (3 weeks vacation, and 10 days of statutory holidays), actually earns $16.60 for each hour they work (assuming they had no other time off the job for illness or layoff). But I have never heard someone adjust their hourly pay in that manner to reflect their entitlement (legal and otherwise) to paid time off.

According to the methodology of all-in hourly labour cost, paid time off (since it reduces the denominator over which all-in labour costs are calculated) directly increases all-in hourly labour costs. Each week of paid time off (including the two weeks of vacation required per year under Canadian law, and the roughly two weeks of statutory holidays also required by Canadian law) translates into a roughly 2% increase in hourly labour cost.

Again, his argument is unconvincing. Although undoubtedly, for example, workers who receive a minimum wage, unionized workers and other workers are unlikely to conceive of paid holidays as part of their compensation, this hardly means that the paid holidays and vacaction do not form part of their compensation. Stanford himself admits this: “Paid time off can be considered a form of compensation.” Actual hours worked and the total amount of compensation received by workers are the relevant statistics for determining variable capital costs and not imputed hours worked during holidays and vacation (when workers are not subject to the direct power of employers).

Furthermore, even some union reps in the auto industry conceived vacation pay and other fringe benefits as part of the compensation package. From Frank Lovell (May-June 1968), “The Reuther-Meany Split,” in pages 36-58, International Socialist Review, page 51:

Reuther accurately reported the new wage scales as follows:”The average production worker will receive a 20-cents-an-hour wage increase upon his return to work plus a three per cent annual wage increase in the second and third years of the contract. These wage increases, together with the impact of the ‘roll up’ factor, will amount to an average of 58 cents an hour over the three-year period of the contract.”The average skilled trades worker will receive a 50-cent-an-hour wage increase upon his return to work plus a three per cent annual wage incrase in the second and third years of the contract. These wage increases, together with the impact of the ‘roll up’ factor, will amount to an average of $1.02 an hour over the three-year period of the contract.” (“Roll up” consists of increases in wage-related fringe benefits such as holiday pay, vacation pay, shift premiums, etc.) [my emphasis]

Let us look at vacation pay and holiday pay. Essentially, it means that workers receive payment without having to work for their employer during that specific time. The amount of labour performed is thereby reduced to that extent than otherwise would be the case, with a flow of money (and indirectly commodities) going to the workers. The wage is not reproduced during that time, and no surplus value is produced either since no labour is performed.

Holiday pay and vacation pay are tied to work performed because those who do not work for the particular employer simply do not receive such holiday pay and (more obviously) vacation pay– but this condition seems too often to be overlooked. The payment of holiday pay and vacation pay is tied to the need for the worker to have actually worked for the particular employer.

The payment is tied to labour having been performed, but not from any labour performed during the holiday or vacation. Since the workers receive the money and not the employer, the money represents the equivalent of higher wages and less surplus value available for the employer.

There is no logical reason why, when calculating the rate of exploitation, vacation pay, sick leave pay and holiday pay should not be included in the calculation.

The issue is not clear cut, but some Marxian works also include vacation pay, etc. as forms of compensation. From Edward Wolff (1987), Growth, accumulation, and unproductive activity: An analysis of the postwar U.S. economy, pages 61-62:

Mean real labor compensation seems the most direct measure of the costs of reproducing labor power. Employee compensation is the sum of wages, salaries, and tips; fringe benefits such as health insurance, pension contributions, vacation pay, and the like;

Shaikh and Tonak also consider them to form part of compensation. From page 304:

We use employee compensation (EC) because it includes wages and salaries of employees as well as employer contributions to social security. This is the appropriate base for estimates of variable capital, since it represents the total cost of labor power to the capitalist.

They reiterate their view on page 322:

Employee compensation being the sum of supplements and wages and salaries.

Impact on hourly cost of layoffs & downtime

Stanford reasons as follows in order to exclude this subcategory from the determination of the costs of employing workers:

The Impact of Downtime and Layoffs Even more far-fetched is the notion that time away from work resulting from illness, layoff, or plant shutdown should also be reflected in your “hourly wage.” Time off due to illness or layoff is not a contractual benefit; it is clearly beyond the control of both workers and their union. Suppose that workers are laid off for 8 weeks in a year because of slow sales. This reduces annual hours worked by 320 hours. That’s a reduction of as much as 20% in hours worked (after adjusting for paid time off) – causing a corresponding increase in the apparent hourly cost of fixed annual benefits (like the pensions, health premiums, and other benefits listed above). Based on the level of benefits described earlier, this amount of downtime (not unusual given recent experience) would add $3 per hour to all-in costs. A longer six-month layoff would add over $10 to the all-in hourly cost!

This seems like a double penalty: first workers experience the income loss and insecurity of being laid off for significant amounts of time. And then they are “charged,” in the form of a higher apparent “wage,” for the fact that they didn’t work for the complete year.

Differences in the number of hours worked account for a significant portion of differences in the all-in hourly labour costs between different companies. Chrysler Canada’s all-in labour cost calculation for 2008 (which has been widely debated in the course of current restructuring discussions) was based on a very low average level of hours worked per worker that year (just 1550 hours). That was significantly lower than the number of hours worked per worker at GM and Ford that year – and far, far lower than average hours worked at Toyota and Honda plants (which until recently have been running flat out). This difference in assumed hours worked accounts for about $2 per hour in all-in labour cost differences between Chrysler and the other two North American producers in Canada. And it accounts for $4 or more per hour of the all-in hourly labour cost differences between Chrysler and the non-union Canadian facilities.

Is a worker really “more expensive” because he or she didn’t work the full year, due to downtime associated with slow sales? Not really. This is not an issue of compensation. This is an issue of capacity utilization – a variable which is clearly a responsibility of management to optimize, and is beyond the control of workers and their union.

The hours not worked do not involve exploitation, and the hours not worked do involve payment without being under the direct control of the employer due to fixed costs, such as health insurance premiums. They are like vacation pay, holiday pay and sick-leave pay.

As for the decrease in hours worked, of course, if the number of hours decreases, with fixed benefit costs, the cost per hour employed will increase for the capitalists in general. This was seen indirectly when I calculated the rate of exploitation of Magna Internaitonal workers for 2020, during the pandemic (The Rate of Exploitation of Workers at Magna International Inc., One of the Largest Private Employers in Toronto, Part Three, Updated, 2020). As I wrote in that post:

The substantial decrease in the rate of exploitation is likely due to the treatment of workers as “fixed costs” as the pandemic forced employers to retain workers despite the relatively extra costs associated with it (partly offset by federal, provincial and municipal supports).

The substantial decrease in the rate of exploitation for Magna International workers was likely largely due to a decrease in the production of surplus value, although there was also a decrease in the costs of workers for Magna International:

2020: So, with the adjustments in place: s=1081; v=2,509. The rate of exploitation or the rate of surplus value=s/v=1081/2,509=43%.2019: So, with the adjustments in place: s=2,258; v=2,862. The rate of exploitation or the rate of surplus value=s/v=2,258/2,862=79%.

Surplus value decreased by 1,177, or 52 percent ((1081-2258)/2258×100); variable capital decreased by 353, or ((2509-2862)/2862×100)=12 percent.

The substantial decrease in the rate of exploitation for Magna International workers is likely due to the fixed costs associated with keeping them on payroll while not exploiting them (because they did not work or perform labour).

Stanford’s justification or logic for discounting the increased costs of workers per hour worked seems to be: because the relation of payments to workers and hours actually worked is independent of the workers and their unions, these increased costs should not be calculated that way, he reasons:

Time off due to illness or layoff is not a contractual benefit; it is clearly beyond the control of both workers and their union.

The rate of exploitation is hardly to be calculated on the basis of what workers and their unions are able to control. The issue is: What does it cost for the workers to produce the value of the labour power or the capacity of workers to perform work relative to the value they produce for free?

I reject Stanford’s reason for excluding this subcategory.

Cost per hour worked of SUB $1-3

Standord reasons as follows for rejecting this subcategory:

Over the years auto unions have negotiated a range of income security programs to protect against the effects of the layoffs (which are regularly incurred in the auto industry due to market swings, new model launches, and other factors inherent to the auto industry). These are called supplementary unemployment benefits (SUB), and they top up the benefits received from public unemployment insurance programs.SUB costs are incorporated into all-in labour cost by attributing them to the hours which were actually worked (by those workers who were not laid off). Are SUB benefits a form of compensation? Yes, in a form. But it is not compensation received by the workers who are still working. SUB benefits are received by the workers who are laid-off (as a partial compensation for the cost they incur as a result of the lay-off). And by far the best way to reduce labour costs, in this context, is to put autoworkers back to work: they enjoy more income and security, the company pays out less SUB costs, and the cost per hour worked of all other benefits declines by several dollars.

Because of the extensive downtime experienced in most auto plants in recent years, SUB and related programs can add $3 or more to all-in hourly labour costs in CAW facilities.

The Canadian government website reads:

Supplemental Unemployment Benefit Program

Overview

Employers can use a Supplemental Unemployment Benefit (SUB) plan to increase their employees’ weekly earnings when they are unemployed due to a temporary stoppage of work, training, illness, injury or quarantine.

It is certainly true that those who continue to be exploited directly by GM do not receive the benefits of SUB-but exploitation is hardly just an affair of temporary changes in the distribution of work such that only those workers who happen to have seniority are exploited throughout the year. Layoffs in the auto industry have occurred often enough for unions to attempt to address the issue. From Robert Albritton (2022), A Japanese Approach to Stages of Capitalist Development: What Comes Next?:

For this reason, the continuation of even the partial commodification of labour-power, requires that the labour market be supported by all sorts of protections, guarantees, and stabilizing mechanisms. The auto industry in general took the lead in this respect with “productivity deals” that ensured wage increases in line with productivity increases and with “supplementary unemployment benefits” (SUBS) that protected worker’s income during the annual lay-offs in the auto industry, and with pension plans, early retirement, medical benefits, etc.

Unions obliged GM to pay laid off GM workers (not fired GM workers) a top-up to unemployment benefits. This is a condition for GM to exploit the collective labour of the particular bargaining unit and forms part of the cost of employing GM workers. To exclude this cost from variable capital would exclude the flow of additional money (and commodities) that temporarily unemployed workers receive from GM. Since variable capital is supposed to measure the amount of labour required to produce the equivalent of labour power as a commodity available on the market, and SUB is a condition for such availability in this situation, it should be included as part of variable capital that GM pays.

One historian has interpreted SUB as resulting in a “greater share of the pie” going to workers. From David Noble (1984), Forces of Production: A Social History of Industrial Automation, page 253:

Reuther embraced wholeheartedly the tempo of technology and the gospel of growth, and sought to halt job erosion through industrial expansion which would presumably raise the level of aggregate demand for labor. At the same time, he strove to secure for workers a larger share of the expanding pie, through guaranteed wage agreements and so-called progress-sharing agreements (as with American Motors), and to ease the plight of displaced workers, through supplementary unemployment benefits, advance notification clauses, and company-financed retraining programs (as with General Motors). [my emphasis]

I also reject Stanford’s reason for excluding this subcategory; SUB forms part of the costs of production of GM workers.

Statutory taxes $3-4

Stanford notes the following:

The all-in hourly labour cost methodology also considers various employmentrelated
taxes paid by employers to governments. In Canada, these statutory
costs include four major items:

Employer CPP premiums (up to a maximum of about $2050 per year

Employer EI premiums (up to a maximum of about $1000 per year)

Employer Health Tax (equal to about 2% of earnings)

WSIB premiums (variable rates, usually about 3% of earnings)These government payments amount to around $3-4 per normal working hour in
Canadian auto plants.

He makes the following comment to justify excluding them from the cost of workers to GM:

These tax payments, while they fund important public programs, obviously do not
constitute compensation for workers.

If by compensation is meant “not received in the current year,” then it is true. However, costs in the present for employers can be deferred revenue for workers; in the case of the Canadian Pension Plan (CPP), Canadian employers must pay the same amount paid by employees, up to a maximum of contributions per year. I fail to see why this should not be included in the cost of variable capital to GM. Furthermore, if pension benefits paid out by GM or deferred government pension benefits are excluded from the current year, when would they ever be included in the calculation of the value of labour power?

Unemployment insurance is a little more difficult to assess (in Canada, the Canadian government euphemestically calls it employment insurance). Some if not the majority of GM workers may not access unemployment insurance benefits at all during their life. It may not ever therefore be a deferred form of revenue for them. However, it should be remembered that GM workers can access unemployment insurance if they are laid off (and they also can receive SUB) and thus the same logic applies as the subcategory for SUB.

(By the way, the unemployment insurance that workers receive should also be included in the calculation of variable capital, but this would require more detailed information about the absolute amounts, the number of GM workers laid off, the average period of time laid off and other data which I doubt is readily available. Perhaps those with more skill or with better access to data could provide details.)

Furthermore, the nature of insurance in general is such that payments or costs may not result in any flow back in the form of services rendered. How many reading this post have purchased insurance of one form or another (such as car insurance or travel insurance) only not to use it? It is a cost that may never be recouped and is grounded in the nature of the capitalist society–which is subject to unemployment in various sectors at different times. Costs paid by employers associated with unemployment insurance is a charge on the class of employers for funding former workers who have had their relationship to any particular employer severed (or who severed it themselves by quitting–although here in Canada workers must prove “just cause” for quitting in order to be eligible and must not have been fired for “just cause”).

For GM workers who are never laid off, unemployment insurance is obviously not a form of compensation for any GM worker–but it is still a cost of reproducing the value of labour power as a commodity (if not the particular set of workers called GM employees). It is part of the cost for GM of being able to exploit this particular set of workers.

The Employer Health Tax, according to the Ontario government website, is:

The Employer Health Tax (EHT) is a payroll tax on remuneration (for example, salaries, wages, bonuses, taxable benefits, stock options etc.) that employers in Ontario provide to current and former employees.

The purpose of this tax is to assist in providing the government with revenue to fund health care in Ontario.

The provision of certain services by the government without workers having to pay for them constitute part of the “social wage” of workers. In this particular case, sufficiently large employers are forced to pay for part of those services (smaller capitalist firms are exempt). These costs for the employer are necessary to ensure a certain level of health services and health of workers. I fail to see why they should not be considered part of the necessary costs that Ontario employers must pay if they are to exploit Ontario employees.

An indirect argument for including the EHT in the calculation of variable capital is provided by Wolff (cited above), page 78:

One further refinement should be added. Not only private consumption but also publicly provided consumption is required to reproduce the labor force. In particular, part of the government’s expenditures on education, health, fire protection, roads, and the like
contributes directly to the welfare of workers. Thus, in order to correctly estimate the necessary consumption of workers, government expenditures on productive goods and services, Gp, must be distributed among the beneficiaries of the expenditures.

As for Workers’ Compensation premiums paid by GM, although Shaikh and Tonak (1994) do not directly address this specific category, they do generally include “employer contributions to social security” in their calculation of the value of labour power. Page 108:

for wages we use “employee compensation” (EC), which includes wages and salaries of employees as well as employer contributions to social security. Employee compensation is the appropriate measure upon which to base our estimates of variable capital, since it represents the total cost of labor power to the capitalist.

Rodrigo Finkelstein (2015) more specifically argues that workers’ compensation premiums constitute the exchange value of workers’ injuries,”The Commodity Form of Safety Information,” in pages 610-623, triple C, Volume 13, Number 2, page 622:

Through injuries, diseases and deaths, workers transfer to the premium the value they themselves lose during the labour process by the destruction of their own use-value. Workers’ use-value—i.e. labour-power—uure [sic–this term indicates that something is quoted exactly as written despite it likely being an error]-appears in the value of every commodity as the premium.

Like shift premiums and overtime, workers’ compensation premiums can be conceived as compensation for accelerated use of workers’ labour power; unlike shift premiums and overtime, the money is not appropriated directly by individual workers but is mediated through a bureuacratic appropriation and distribution process that pools the accelerated use of workers’ labour power at the provincial level in Canada.

(These premiums should not, however, be considered the accurate costs of accelerated use of the labour power of workers. Premiums are based on claimed employer-dictated work-related diseases, injuries and deaths, but actual employer-dictated work-related diseases, injuries and deaths is much higher (see Working and Living in a Society Dominated by a Class of Employers May Be Dangerous to Your Health).

It should come as no surprise that Stanford ignores workers’ compensation as a cost. Firstly, he considers the view that what is healthy for “the economy” is somehow also healthy for workers. This correspondence may to a certain extent arise because capitalists, ultimately, must rely on human bodies if they are to exploit them, and unhealthy bodies may be detrimental to their exploitation.

On the other hand, there obviously is a counter-tendency for employers to create working conditions that are dangerous for workers in one way or another (see my critique of Stanford’s attempt to treat the capitalist economy as if it were an economy primarily based on the production of products that workers, citizens, immigrants and migrants need–rather than a capitalist economy designed to obtain as much surplus value as possible–at the expense of workers in the posts Economics for Social Democrats–but not for the Working Class, Part Three: The Health and Safety of Workers and an Economy Dominated by a Class of Employers Are at Loggerheads and Economics for Social Democrats–but not for the Working Class, Part Four: Is There Such a Thing as a Responsible Employer in Relation to the Health of Workers?).

Stanford’s analysis of the real cost of the value of labour power matches his economics for social democrats. His economics of capitalism for “everyone” (the main title of a book he wrote) is really an economics for social reformists–and is hardly expressive of the interests of the working class.

Stanford’s dismissal of workers’ compensation premiums as part of the cost of the value of labour power reflects his social-democratic views. Workers’ compensation premiums are linked to the determination of the value of the loss of various parts of the body, for example–equating money and the loss of human body parts. As Nate Holdren (2020),in Injury Impoverished: Workplace Accidents, Capitalism, and the Law in the Progressive Era remarks, page 5:

I remarked that it felt a little creepy that my hand had a dollar value. The lawyer laughed and agreed that it was creepy. He told me that there were tables that listed the value of all the different body parts. … I repeated that it was a creepy idea that my body parts in particular had a dollar value, and that in general there were tables written down with the value of body parts calculated in advance. The lawyer replied that a lot of people got hurt at work and that the injuries and the payments for them were all a regular process.That meeting with the lawyer is where I first encountered what I now think of as the “tyranny of the table,” but it is both more and less than tyranny. What I mean by the tyranny of the table is that within compensation laws human lives and human suffering have the fixed monetary values ascribed – no more than that, and not subject to discussion. What doesn’t fit into the values of the table? Nearly everything. All of the elements of a human being other than our paychecks.

Behind the numbers lie real human beings, with histories linked to other human beings in various ways. Workers’ compensation itself hides this reality behind the numbers, and Stanford’s cavalier dismissal of the payment of premiums by employers as part of the cost of the value of labour power reflects his own dismissal of the real and necessary experiences of many workers working for employers–and the diseases and injuries they experience that they suffer–and the deaths (the ending of any possibility of further human experience or any possibility of further human progress).

In Stanford’s haste at being a “progressive economist,” (he is the founder of the Progressive Economic Forum), he dismisses too hastily various costs that are relevant for characterizing the experience of workers in a capitalist society (such as the premiums paid to the Workers’ Compensation Board).

Although this post is about the rate of exploitation, it should never be forgotten that this rate of exploitation is linked to real people being used as things for the benefit of employers, with real negative consequences for members of the working class. Holdren points out how workers who work for an employer are often under the threat of being injured in one way or another, and if they are, their lives are often changed forever, page 1:

Nettie Blom worked in the laundry of a hotel in Yellowstone Park. On June 30, 1900, Blom was operating a machine called a mangle, which used steam-heated and steam-powered metal rollers to iron flat linens. The wet cloth stuck to her hand for a moment too long, and she was pulled into the machine. Blom’s hand was crushed and burnt. When a co-worker managed to free her from the machine, Blom’s hand looked like “boiled meat.” Three of her co-workers fainted at the sight. Blom suffered terrible pain and lost the use of her hand due to her injuries.

Stanford’s dismissal of workers’ compensation premiums as part of the value of labour power also hides the shift from what Holdren calls the tyranny of the trial to the tyrranny of the tables. The (unlikely) possibility of workers suing and winning a case against their employer constituted part of the tyranny of the trial historically, but gradually governments shifted the issue of health and safety compensation from the courts to government bureaucracies–Worker’s Compensaiton Boards. This shift from the workers’ financial point of view had the advantage that compensation would be forthcoming for proven injuries, disease or death–but it had the major disadvantage of eliminating any exposure of the real human damage and suffering that is so often connected to workplace injuries, diseases and death–an exposure that was at least minimally possible during a trial. In its place arose what Holdren calls the “tyranny of the table,” which depersonalized human injury and suffering at work. This is an important issue that the social-democratic left simply ignore or sidestep through the use of such euphemisms as “decent work,” “decent jobs,” “fair contracts” and the like. It is appropriate here to quote Holdren here more extensively about some of the implications of what the paying of premiums has involved for silencing workers’ grievances. Pages 115-118:

MORAL THINNING AND IMPOVERISHED INJURYIn order to standardize payments and thus create predictability for employers, compensation laws removed from the law arguments about injustice and narration of the individual effects of injury. This loss of deliberation changed the ethical grammar of the law, so to speak, deepening the eclipse of recognition, further impoverishing injury. The human meaning of injury had no place in the law. I call this phenomenon moral thinning: from murder to statistics. Non-financial harms also had no place under compensation laws. Pain and loss became newly worthless as the law provided no more space for people to narrate what it meant to lose a limb or a family member in an industrial accident. Injured wage earners became conceptually disembedded from their social and interpersonal contexts.

There is an element of moral thinning involved whenever the commodification of persons begins to occur, because commodification must ignore differences and particularities, setting aside whatever is unique or nonequivalent about them. Commodification tramples on singularity. This makes no difference when singularity makes no difference: the uniqueness of my morning cup of coffee does not matter; what matters is its instrumental use in my struggle toward wakefulness. The uniqueness of human beings, however, does matter: the reduction of human beings to abstract instrumental objects should trouble us. Recognition and commodification co-exist at best uneasily.

The moral thinning of injury under the tyranny of the table is more apparent when juxtaposed to the tyranny of the trial. Despite the many limits of the court-based system of employee injury law, that system did allow some space for fragments of the experiential truths of injury, which made possible elements of justice as recognition. As historian Kimberly Welch has put it, “[s]torytelling is omnipresent in human discourse.. . . Telling stories in court is an attempt to organize, interpret, and direct the world in which one lives, and the stories told in adversarial processes signal the narrator’s interpretation of how the world ought to operate.” The contending oughts embedded in legal stories made courts into places of normative deliberation, places where the contest of stories had explicitly moral and political stakes.

Access to that site of deliberation, and the recognition that came through that access, is likely part of what working-class people wanted from the court-based system of employee injury law. As historian James Schmidt has put it, injured plaintiffs and their families “came to court with a desire to talk about the miseries that had befallen them.” That telling intersected with other actors in court to produce what Schmidt calls “judicial morality plays.” Going to court was one kind of ritual through which people processed and, in important respects, produced the meaning of what Schmidt rightly calls industrial violence. There was, then, some space for this ritual use of law under the tyranny of the trial. With compensation laws, employee injury law was deritualized, no longer made available to working-class people in the same way.

To be clear, compensation laws never said that no other framework for valuing human beings existed in society, but these laws did not allow any other such framework to touch the legal response to employee injury. In the court system multiple systems of valuation could intersect, while under compensation laws non-pecuniary valuations of people, their experiences, their relationships, and their bodies had no legal space. The point is absolutely not to celebrate the tyranny of the trial, but to use the courts’ narrative and value plurality to highlight the moral thinning of injury under the tyranny of the table. In the court-based system of injury law at least it was possible to pose the questions of whether or not an injury was a wrong, and what it meant in the lives of the persons affected. There was no more space for these questions or for the answering stories of injury and its effects under the tyranny of the table.

Workers’ compensation premiums serve in part to hide the viciousness of a society dominated by a class of employers–a viciousness hidden by such social-democratic phrases as “decent jobs,” “decent work,” “fair contracts” and other euphemisms accepted by many on the so-called left these days. After all, they imply, working for an employer is not really all that bad; such is the moral thinning of social democrats these days. This is a class cost–Stanford’s dismissal of it notwithstadning.

Returning to the issue of the cost of workers, this cost of $63, of course, is probably less since GM workers in other parts of the world (such as in Mexico) would receive substantially less. However, without access to such detailed statistics, I will assume that the $63 per hour is still the average hourly wage for GM workers; perhaps Mr. Stanford (and Mr. Gindin) could provide more detailed statistics. Such statistics would be most welcome.

Given a wage of $63 U.S. an hour, and given an estimated 1,800 hours of actual work per employee (see page 2 of Stanford’s article), and given 164,000 employees, the result is:

Total wages and benefits: $18.5976 billion

Calculation of the Rate of Exploitation

We now have sufficient information to calculate the rate of exploitation of GM workers (in billions of U.S. dollars)

Adjusted Income before income taxes: $7.383 billion=s
Total wages and benefits $18.5976 billion=v