Member Labor:

If your co-op depends to any degree on member labor, you are naturally concerned about the effectiveness of that aspect of the co-op's operation. But the complexity of the roles people play in a cooperative -- the overlapping roles of owner, customer, member worker, employee, director -- can make measuring effectiveness seem difficult. In addition, while effectiveness results from a combination of methods and design, at the same time the decided mission of the co-op (or the lack of a decided misssion) has a major effect on those designs and methods.

An example of how mission, methods and measures are related is the way that different organizations deal with human resources. On one hand, there can be the assumption that if a big gap is measured between the best performer on a job and the average or below average person, that's because the best person is a genius and the lesser performer is a turkey. The response? Weed out the turkeys and find more geniuses. After all, it's more expensive and difficult to teach turkeys to be geniuses, and we must operate cheaply in order to survive. Sound familiar?

Some may say "Let many flowers bloom"; yet some of us have learned the hard way that we have planted certain exotic species in hostile environments, and have come to question our assumptions. Particularly with member labor, there appears to be confusion between commitment to member labor as a mission, and commitment to it as a method.

By contrast, there can be the assumption that the best performer differs from the average mainly because he or she is good at figuring out ways around all the unnecessary barriers, self-defeating disincentives, and generally silly obstacles that organizations place in the way of people's performance. An organization that places an explicit priority on the reduction of waste or on organizational development will tend to deal with worker performance problems by looking first at the design of the job, incentives and disincentives to perform well, consequences of good and bad performance (or the lack of them!), and the learning resources available to the worker. Only when these cannot be improved upon would a worker be transferred or fired.

One of the problems with member labor has been the difficulty of finding effectiveness measures and describing operating methods for member labor in co-ops. We've seen articles in which different people, each using consistent logic, quote numbers and conclude that member labor is a delusory scam, more expensive than paid staff; or that member labor is a great bargain for both the member and the organization. Naturally, given these differences and given my point about how methods and measures are shaped by purpose, we should expect to find disagreement about the mission of co-ops and member labor. Some people feel "burned" by the member labor experience, or never have relied on member labor at their store, and wouldn't think of putting it in their mission statement. Yet a number of stores have stated that they regard member labor as an important part of their mission: member labor is seen as a necessary expression of the purpose and identity of the cooperative.

-

Co-ops discussed here:

Cape Ann Food Co-op: Gloucester, Massachusetts. A patron operated store until 1982, when member labor became optional; opened to non member sales in 1985; current sales around $1.2 million annually; current membership of around 700 households comprising some 1100 individuals.

Duck Soup Coup: De KaIb, Illinois. A patron operated cooperative store founded in 1974; annual sales currently around $400,000; membership around 800.

Durham Food Co-op: Durham, North Carolina. A patron operated store until 1986, when member labor became optional. Current sales around $340,000 annually; membership is 1000+. Park Slope Food Co-op: Brooklyn, New York. A patron operated store since 1974; current membership 1900+; annual sales approaching $2,000,000.

Performance comparison

Are we about to resolve the issues of mission and measurement? Hardly. But our co-op in DeKaIb, Illinois, has recently been dissatisfied enough with our performance and our sense of mission to take a fresh look at member labor. We did so from a perspective often used by training managers: we arrived at a list of performance measures and went looking for the "exemplary performer" -- the organization doing the best job according to these measures. We wanted to analyze the contingencies and methods used that made the exemplary performer the best. Specifically, we looked for co-ops that were the most committed to member labor, the ones that require it of their members: patron operated stores.

Duck Soup in DeKaIb is a patron operated store, one that serves members exclusively, and operates as a kind of direct charge cooperative in which the membership unit is assessed a monthly work stint instead of a fee. (Direct charge cooperatives, numerous in Canada but rare in the U.S., sell to members only and require a fixed periodic -- biweekly or monthly -- fee or fees to cover capital and operating expenses; the fixed fee is an incentive for members to purchase more of their groceries at the co-op.) Duck Soup was compared with three other stores we found that have met the criteria we were looking for:

• A full-time staff;
• A capital plant using commercial grocery equipment;
• An inventory (as opposed to a preorder system); and
• An operation that serves working patrons exclusively.

The other stores were: Cape Ann in Gloucester, Massachusetts; Durham Food Co-op, NC; and Park Slope in Brooklyn, NY. Of the four, only Park Slope and Duck Soup continue as patron operated stores; Cape Ann and Durham both eventually abandoned the system: in 1986 at Durham, in 1982 at Cape Ann. (Hence, the years from which these co-ops' performance data were taken ended with those dates.)

In examining the member labor systems at these four stores, the measures we looked at were:

• The amount of member work contributed, in hours.
• The productivity of staff labor, in sales per full-time worker.
• Staff costs, as a percent of sales.

We also were interested in such work structure methods and outcomes as:

• The type of supervisory roles and training techniques used.
• For a given work objective, what percentage of the work did members do, and how much did staff do?
• Degree of fit with our "exemplary co-op" model.

The exemplary co-op

Our working definition of the exemplary co-op included a consideration of three sets of needs: member needs, organizational needs, and staff needs. We considered that members have need for economic, cultural/social, and educational benefits. The organization has need for effective management and sufficient investment, while the staff have need for job satisfaction/development, and competitive compensation related to their performance. The "exemplary co-op" would make substantive contributions to all these areas at once, with equal priority placed on each.

The co-op from our sample that came closest to the exemplary model is Park Slope. Members have indicated their satisfaction with the benefits of the coop by voting with their time and their wallets: the co-op is approaching the two million dollar annual sales mark with strong growth, involving 1900+ people. They work three hours per four weeks per adult per household. In other words, a two-adult household must contribute six hours of work per four week period, or neither person can belong. This is a substantial commitment, made by people receiving substantial economic and social rewards from their organization.

Looking over the organization generally, we could find no fault with basic indicators. Investment in the co-op is sufficient to reasonably support Park Slope's operations, and staff salaries and benefits are comfortably above the "staff labor subsidy" level. The staff are longtime employees who appear to be highly motivated and good performers.

On the measures that most interested us, the current staff cost percentage of sales is the lowest achieved of the four, and quite low by national standards: about 7.5 percent. Member labor input is also the highest of the group: from 2 to 4 times higher than the others. This record has been sustained for many years, again indicating powerful incentives and a well organized system. Productivity has been averaging near $400,000 of sales per full time equivalent, an astronomical figure by conventional national standards and from 40 to 400 percent more than the peak rates achieved for the other three stores studied.

Structure, method, results

Importantly, we found large differences in productivity between systems using different methods to structure the work. In the two systems with the lowest productivity, members could work at a different time slot each cycle if they so desired. These systems were driven by the convenience of members in picking time slots. The amount of help available to staff in getting a task accomplished at a given time was subject to fluctuating sign-ups as well as to no-shows. Also, training was largely learning-by-doing with individualized verbal instruction mostly delivered by staff people.

In the two higher productivity cases, the labor systems were and are driven by scheduled tasks. Highly defined tasks with specified criteria for performance would be scheduled for specific time slots, and a certain number of work slots deemed necessary to complete the task. Members would be offered these work slots as permanent positions -- a kind of "microjob" treated much like any wage position is. Members could only join if there was a job available for them to commit to.

In these more productive systems, oriented around scheduled tasks, the goal is to turn whole tasks over to members, with staff providing the roles of job designer, support person, and emergency backup. Training materials in the form of written instructions, and job aids such as checklists, illustrated procedure charts, and special forms are used to refresh memories and allow peers to train newcomers.

The first system assumes variable sharing of tasks between members and staff people and places extra burdens on staff to cover shifts when help is less predictable and need for individualized attention is greater. The second system supports member work with written training materials and peer instruction first and staff backup second. Given comparable work requirements, the task oriented system with specified performance, criteria, and conditions for task performance was up to twice as productive as the member convenience oriented system.

It appears to be no coincidence that the two co-ops with the lowest productivity found themselves overwhelmed by conditions when their growth curves leveled off, and they attained their peak market share as patron operated stores. In the two cases where this happened to date, sales and membership declined; staff reported that members increasingly felt that the co-op's benefits did not justify the effort. Also, some staff felt snowed under by the impossibility of keeping up all the "personnel" functions of member work status, training, supervision, plus the sharing of clerking tasks, while also trying to accomplish managerial tasks necessary to establish the store solidly in its market niche.

As mentioned earlier, both Durham and Cape Ann have abandoned the member labor requirement and increased their member equity requirement. Since that change, both stores have seen turnaround efforts rewarded with sales levels 50 to 100 percent higher than their peak achievements under the patron operated system. In both cases, less than 2 percent of members now participate in vestigial member labor programs. In the case of Cape Ann, there are now 15 to 20 percent more members participating in the current equity program than there were members participating in the labor program at its peak, and their numbers are growing, not declining.

Duck Soup has a member labor system which very closely parallels Park Slope's and which has given it consistently superior productivity relative to the other two co-ops studied. Nevertheless, Duck Soup has recently seen its productivity decline severely, reflecting an inability to achieve and sustain sufficient member benefits and sales volume in its much smaller community. It appears to be beginning the same negative spiral foreshadowed by its counterparts. This has already caused, among other things, difficulty with capitalization, and controversy over the co-op's mission.

Conclusions

Patron operated co-ops in the specialty foods market appear to need high productivity levels, and there appears to be only one example nationally of stable success for such a store. It's worth noting that the system installed so successfully in Brooklyn is very hard to replicate outside of a large city with high population density and easy access to resources such as a terminal produce market and competitive wholesale grocers. Co-ops in other communities have found it extremely difficult to put together sufficient incentives for sustained member participation. One attempt to install a close clone of this exemplary model in a much smaller and different market area has had difficulty breaking out of its entrepreneurial phase, in spite of comparable structures. And at least some of those stores that operated without comparable methods and measures proved unworkable much earlier.

More typical co-op stores that have member labor programs often see as many as 10 to 20 percent of their members working in the store -- there are a considerable number of programs that involve scores of people. Many of these appear to use random workshifts task sharing between staff and members, and more reliance on the selection of "qualified" individuals rather than on good job design, written instruction and job aids that allow more individuals to be effective. Consequently, there is overreliance on staff to provide verbal instructions and individualized supervision.

It is the experience of patron operated stores that such a methods package is far less productive than the performance-based systems described earlier. Programs without strict reliance on well defined, scheduled tasks, supported with print-based instructions and job aids, and with careful measurement of staff time used in all administration are undoubtedly wasting paid supervisory time. How many co-ops with such programs know what it's costing?

The goal of any good system of people performance, and the goal of job design and the attendant measurements of performance, should be empowerment. The participators should be as independent, knowledgeable, and as in-charge of whole tasks and functions as possible. This is as true for full-time employees as it is for those who appear to play a very narrow role for a few hours a month. Such outcomes do not automatically follow from placing willing workers in the proximity of experienced people. Rather, such outcomes must be engineered through job design and training support.

Some may say "Let many flowers bloom"; yet some of us have learned the hard way that we have planted certain exotic species in hostile environments, and have come to question our assumptions. Particularly with member labor, there appears to be confusion between commitment to member labor as a mission, and commitment to it as a method. In the case of patron operated stores, member labor has failed as a sustainable method due to mismatch between community and system, plus lack of productive methods -- not due to lack of commitment to the concept.

Members should work in the store because they know that their store benefits thereby, not merely because it's a good deal for them individually. But the benefits to the store cannot long be true unless there are comprehensive cost and result measurements and effective methods in place to validate the members' efforts. Have a look at your own mission, method and measurements. Compare your member labor program to exemplary performers, and look for ways to engineer better performance. In a cooperative, such effort is to everyone's benefit.