Why are coops so rare? If worker-owned businesses are so much better than traditional businesses, why aren’t they more common? Worker-owners face a number of external barriers that make it harder to start and run a cooperative business, a lack of ready access to capital being perhaps the biggest barrier, and as advocates of worker-ownership we usually assume that it is these external barriers, built into the structure of the capitalist economy, that account for the rarity of worker-ownership, both now and in the past.
But looking back over the history of worker-ownership, we see repeated cycles of growth and decline: in different times and places, worker-ownership growing, spreading and then fading away. Why doesn’t worker-ownership ever just catch on as a movement and continue growing? Are the external factors enough to explain these cycles? Are there perhaps some internal factors that also make worker-ownership more difficult? Is there something specific about worker ownership itself that makes worker coops less successful over time?
These are hard questions for us to ask, but if we want our worker-owned businesses to succeed, we need to ask them. We have to be honest about any potential costs or barriers we may face, so we can plan for them. Recently, in 2015, Saioa Arando, Monica Gago, Derek C. Jones, and Takao Kato published a research paper that may start to answer some of these questions. In this paper, the authors ask a simple question of their own: are worker-owners happy in their jobs?
The authors use one of the huge Mondragon coops, Eroski, as their research subject. As we have seen previously, many of the Mondragon coops have seriously degenerated over the years, and Eroski is no exception. Currently, most Eroski employees are not worker-owners. At the time of the research, the Eroski parent company owned 814 supermarket stores with three different levels of worker-ownership: “1) cooperatives with significant employee ownership and voice; 2) cooperatives with modest employee ownership and limited voice (known as GESPAs); and 3) conventional stores with no employee ownership,” (398) and while it is disappointing that Eroski has degenerated so far, the situation allowed the authors to use Eroski as a sort of natural experiment, comparing the performance of worker-owned supermarkets against GESPAs and conventional stores on a number of measures.
They found, as expected, that the worker-owned stores often out-performed the GESPAs and conventional stores, and also, that cooperative members enjoyed 100% job security and wages at least 20% above the industry standard, but when they looked at job satisfaction in the worker owned stores vs the GESPAs, they found that job satisfaction was lower in the worker-owned stores, in spite of the fact that the work was well paid and more efficient. What was going on here?
The authors found a clue to the difference when they looked at how job satisfaction changed over time and in respect to the overall health of the stores. They found that job satisfaction in the worker-owned stores went down when the performance of the stores went down. Based on the data, the authors present two interpretations. First, they suggest that being a worker-owner may be particularly stressful, that workplace democracy places an additional cognitive load on coop members that makes the work harder:
These co-op stores are more efficient and CO-OP workers have bigger financial stakes and voice, and also receive higher wages; however, employee owners with high stakes in the firm are expected to go beyond routine work and to engage in a variety of problem-solving activities. (420)
They also suggest that coop workers may have higher expectations for their work, and are disappointed when those expectations don’t always match reality. Worker-owners may be particularly stressed and disappointed when their coop is not performing as well as hoped:
stress is highest in cooperatives when sales growth is relatively weak […] by being significant stakeholders, CO-OP workers at Mondragon probably expect more from their work, resulting in high expectations and a higher likelihood of disappointment. Such workplace disappointment may be particularly acute when their hard work does not result in performance improvement. (421)
Like with any natural experiment, we have to be careful about the data. Eroski presents us with a complex picture: we have three types of stores with different levels of ownership, and even in the “worker-owned” stores, not all of the employees are themselves worker owners. Further, the authors found that working conditions were generally better for all employees of Eroski, suggesting perhaps that GESPA workers and non-owner workers might have been, to a degree, free-riding off the advantages of working for a cooperative parent company without having to take on the extra cognitive load that comes with running a worker-owned business.
Also, this is just one study, and these results would need to be replicated in many different situations before we could say that, in general, worker-owners are less satisfied in their jobs than non-owners. It is significant then that just last year Imanol Basterretxea and John Storey published a paper that repeated some of this work, but from data drawn in 2011, after the financial crisis, and they again found lower satisfaction amongst employee owners in comparison to GESPA workers, but this time they also found lower satisfaction amongst employee owners in comparison to non-owners and to temporary employees.
So if this is true, what does it mean for worker-ownership as a business model? I think anyone living in the UK or the USA just now would agree that democracy can be hard work. As I follow the struggles for democracy taking place in Hong Kong, I am so glad that I have the privilege of living in a democratic country, but at the same time, there is no doubt that democracies place extra cognitive load on their citizens, particularly when those democracies are under stress, as it is in the UK under Brexit, or in the US under Trump.
I don’t believe that these two facts are actually in conflict: democracy is a great privilege, but at the same time, democracy can be hard work. This is true for political democracy, and is probably also true for economic democracy. Democracy has real advantages, that is clear, but it also has a cost, and it is not surprising if this cost can sometimes be a drag on the performance of a worker-owned business, perhaps even contributing to degeneration over time. We may be advocates of worker ownership, but we can’t afford to be starry-eyed about it. Everything has plusses and minuses. I would argue that the plusses of workplace democracy far outweigh the minuses, but we need to be clear about the minuses if we want to grow worker-ownership as an economic model.
At a practical level, I am not sure if we have a clear picture yet of how to plan for the extra costs of economic democracy, but Basterretxea and Storey’s paper contains some hints. They interviewed managers at Eroski and some interviewees pointed to a breakdown in cooperative culture over time as one reason that worker-owners were less satisfied. They suggested that the success of the worker-ownership model was not made clear enough to members in the good times, so that members would remain committed to the model when a dip in the economy required longer working hours and lower pay:
engagement deteriorated over time; the increased workload for members following the 2008 crisis; and there had been a failure to build on the previous long period of success in order to foster a culture of ‘pride of belonging’. (308)
This is further evidence that creating and maintain a strong cooperative culture in a worker-owned firm is a key to success, but there may be other factors that influence the impact of the cost of democracy on a worker-owned firm’s performance. Clearly more research is required. Stay tuned!
Arando, Saioa, et al. “Efficiency in employee-owned enterprises: An econometric case study of Mondragon.” ILR Review 68.2 (2015): 398-425.
Basterretxea, Imanol, and John Storey. “Do employee‐owned firms produce more positive employee behavioural outcomes? If not why not? A British‐Spanish comparative analysis.” British Journal of Industrial Relations 56.2 (2018): 292-319.