Economics for Democratic
Spring Quarter 2009
• Rather than handouts or library reserve, we will
rely on Wikipedia readings for this week. Three
(short) ones are added to the original two.
The Social Economy
• A number of economists in Europe, Canada
and Latin America -- and a few in the US --
use the term “the social economy” to refer
to not-for-profit corporations and
• Most of the focus today will be on
cooperatives, but a few comments on not-
for-profits will be needed.
• Nonprofit corporations and Non-Governmental
Organizations (NGOs) generally refer to similar
• They are not prohibited from earning profits but
are not permitted to distribute their profits.
• Instead, they are required to use profits and other
resources at their command to advance their
mission, which generally is stated in the charter.
• Thus, I prefer the term “mission-driven
Some Noted Nonprofits
• Drexel University
– Mission: higher education -- “I want them to have good
• The Barnes
– Mission: promote Dr. Barnes’ ideas on art.
– Mission: promote the arts and sciences of domestic
• Museums, symphonies, many hospitals, etc.
• A Mission-Driven Enterprise is controlled by a
board of directors or trustees.
• It is their responsibility to assure that the
organization operates in accordance with its
• If they fail to do so they may be guilty of failure of
trusteeship, a criminal offense.
• Many MDE’s depend on gifts, and that can keep
them on the strait and narrow path.
• A cooperative (also co-operative or coöperative; often
referred to as a co-op or coop) is defined by the
International Co-operative Alliance's Statement on the Co-
operative Identity as an autonomous association of persons
united voluntarily to meet their common economic, social,
and cultural needs and aspirations through a jointly-owned
and democratically-controlled enterprise.
• In other terms it is a membership organization that operates
an enterprise and with membership open to those who have
a particular (non-owner) relation to the organization.
Some Types of Cooperatives
• Worker cooperatives
– Membership is open to employees
• Consumer cooperatives
– Membership is open to customers
• Financial mutuals
– Membership is open to depositors
• Farmer cooperatives
– Membership is open to farmers as raw material
producers and supply buyers.
• The primary significance for this course is the democratic
governance of worker cooperatives -- production carried
on by a free association of the workers.
• Recall the sense of “democratic” that I endorsed in the first
week: “In a democratic system, any person who holds a
position of authority is responsible to those over whom the
authority is exercised.”
• Worker cooperatives are democratic in that sense. Other
cooperatives and profit-seeking firms are not.
• This is not to say that other cooperatives have not made
• Historically, there have been some conflicts between
advocates of worker and consumer cooperatives.
• In game theory we distinguish between “cooperative” and
“noncooperative” solutions. A familiar example is the
• In a “cooperative solution,” the players may form a
“coalition” to choose a joint strategy for mutual benefit.
• There is a rough parallel. Cooperatives of all kinds are
based on the principle of joining together for mutual
• This should not be pushed too far. Cooperative game
theory is too abstract to tell us much about cooperative
enterprises -- and cooperative “coalitions” are not
• 1st Principle: Voluntary and Open Membership
• 2nd Principle: Democratic Member Control
• 3rd Principle: Member Economic Participation
• 4th Principle: Autonomy and Independence
• 5th Principle: Education, Training and Information
• 6th Principle: Co-operation among Co-operatives
• 7th Principle: Concern for Community
– --International Cooperative Alliance
By the way,
• Here in the USA and elsewhere, worker cooperatives have
often come about through the bankruptcy or closure of
• One example -- here in the Delaware Valley -- was the
O&O supermarket chain, based on stores that A&P closed
as unprofitable in the 1970’s when A&P was in
• They continued to operate into the 1990’s -- but are largely
• So cooperatives can sometimes survive where capitalist
firms cannot (why?) But it doesn’t always work.
A Halfway House
• Most large German corporations are required to
adopt “mitbestimmung” -- codetermination --
either with parity or near parity.
• Codetermination with parity means the employees
elect half of the board of directors.
• In some smaller companies, the minimum is one-
third worker directors.
• This seems a highly successful system.
• Most worker-coops are owned by their members.
But worker-ownership may not mean cooperation.
• Employee Stock Ownership plans are fairly
common, but usually prevent any worker
participation in firm governance -- somewhat the
opposite of cooperation.
• However, in a few cases, coops have been based
on worker ownership via ESOPs.
• For economic theory, the broader distinction is between
labor-managed, profit-seeking, and other kinds of
• Labor-managed enterprise=LME.
• A conventional corporation is managed by a “court of
directors” (A. Smith) elected by investors and that is
supposed to direct the enterprise in their interest.
• An LME is managed by a “court of directors” elected by
the employees and that is supposed to direct the enterprise
in their interest.
• A worker cooperative is (ideally) a worker-owned LME;
but in practice there may be non-member employees.
Worker Coops as LME’s
• Historically, many worker coops have been
established that are very near approximations of
LME’s, and the evidence from observation of
these enterprises is a basis for a judgment of the
workability of an economy founded on LME’s.
• Many of those worker coops have been long-
lasting and successful by any reasonable standard.
• Several will be considered in the following.
Yugoslav LME’s 1
• Yugoslavia broke with the Soviet Union in about 1948,
and a few years later, beginning in 1952, began to build a
“superior” socialist economic system, based on labor
• This was probably originally meant to be mostly show.
Certainly the state retained a major role in the 1950’s.
• One-party Communist rule was also maintained.
• However, the workers’ councils and market allocation
became the main economic forces by 1960.
• Subsequent rounds of “reform” to “perfect” labor
management were probably more disruptive than helpful.
Yugoslav LME’s 2
• On the whole, the Yugoslav economic system was pretty
• However, it was export dependent, quality was poor (the
Yugo was notorious) and after 1980 dependent on exports
to eastern Europe and the Soviet Union.
• The collapse of those countries in turn depressed the
Yugoslav economy (as it did those of Sweden and
• Specifically Yugoslav problems -- hatred and wide
contrasts of economic development between the
nationalities -- were always a handicap to the Yugoslav
Yugoslav LME’s 3
• The Yugoslav Federation collapsed in 1991, and much of
its economy was destroyed in the wars that followed.
• “Privatization” largely took the form of turning over
formal title to the workers themselves.
• In Slovenia and Croatia, the more industrialized regions,
there have been efforts (in the enterprises themselves) to
retain labor management, although this is inconsistent with
European Community regulations and the government has
pressured them to eliminate it.
• Anyway, the first economic theory of LME’s was based on
the Yugoslav system and was initiated by a paper of
Benjamin Ward. (1958)
Ward’s Model Simplified 1
• Maximize per-worker (or average) dividend
• dividend is revenue minus cost, not
including labor cost
• If Q is the quantity of output
L is the quantity of labor
K is the quantity of capital
Ward’s Model Simplified 2
• Then, if labor and capital are the only two inputs, the
production function (which shows relationship
between output and inputs) is:
Q = f(L, K)
• In the short run, capital (K) is fixed
• Suppose that the enterprise pays a “rent” or total cost
(T) for the capital it is using
– since K is fixed in the short run, T is fixed
• So if we let Y = total dividend and
P = the price of the output, then
Y = PQ - T
Ward’s Model Simplified 3
• This equation says that dividend is revenue (PQ) minus the
total cost of capital
• a more complicated (i.e., more realistic) production
function would include more inputs (like fuel,
intermediate products, etc.) but would not change
the basic result we are after here
• Per-worker dividend (which Ward’s model will
maximize) is just Y divided by L
Y/L = (PQ - T)/L
Ward’s Model Simplified 4
• A little wizardry (i.e., calculus) shows that when
per-worker dividend is maximized the following
VMPL = (PQ - T)/L
– where VMPL is the value of marginal product
of labor = P*MPL P * L
Relationship Between VMP and
• If VMPL>Y/L,
then adding an
additional unit of
labor will raise
• If VMPL<Y/L,
then reducing L
will raise Y/L
Comparison With Capitalist Firm
• Assuming profit (J) maximization
J = PQ - wL - T
where w = wage
• At the profit maximizing employment of labor
VMPL = w
• Value of labors’ marginal product equals wage
• This is a short run analysis. As Ward notes, the conditions
for long run equilibrium are identical to those for a
• J. Vanek built an extensive economic theory of LME’s and
LM economies on this.
• It has been criticized on the grounds that the assumptions
• They are -- but they are equally unrealistic in their
application to capitalist economies.
• The correct implication is that all arguments that markets
lead to efficient resource allocation are unsound.
• The key assumption is that capital is supplied by
“perfectly competitive capital markets.”
– Horizontal supply curve of capital to the firm:
– No such thing has, ever will, or can exist in the real
– It would imply infinite leverage.
• In the real world, the allocation of resources
depends on the distribution of wealth.
• No neoclassical nor Austrian economic theory
allows for that.
Arguments in Favor of LMEs
• Eliminates capitalist dichotomy between
management and labor
• Greater social justice in distribution of income
– distribution according to decision of the
Criticisms of LMEs
• Misallocation of labor. Cooperatively managed firms tend
to hire less labor.
– efficiency requires equality of VMP in all uses
– if VMPL = w and all firms face same wage, then all
firms’ VMPL will be the same
– no such mechanism to equate VMPs in market
socialism. If VMP1 > VMP2 then greater value can be
produced by reallocating labor from Enterprise 2 to
Further Criticisms of LMEs
• We can show that the cooperative monopolist
restricts output (i.e., misallocates resources) even
more than the capitalist monopoly
• Labor managed enterprises may underinvest in
capital in the long run because workers will drain
firms of extra income in the short run, especially
the lack of property rights in firm assets reduces
their incentive for supporting long horizon
Coop Experience as Evidence 1
• In the 1970’s a group of economists formed around Vanek
with an interest in research on LME’s.
• When we learned about the century-plus of experience
with worker cooperatives, some of the group did empirical
research on enterprises of that kind.
• The most intensively studied cases were
– US Pacific Northwest worker-owned plywood
– A complex of enterprises at Mondragon, Spain.
• Other important, but less-studied instances include
Legacoop, Northern Italy, especially Emilia Romagna,
experiences in several other European nations.
Preliminary Results 1
• The Ward model was not supported, at least in the
simplified form given here.
• There is also little or no evidence of underinvestment.
• However, the enterprises typically are undercapitalized
because of inability or unwillingness to raise capital from
• Worker cooperatives show differences in form and legal
definition that imply differences at least where incentives
to invest are concerned. There may be no “general” model.
Preliminary Results 2
• Another (hypothetical) criticism from opponents
of LMEs is that labor productivity would be very
• The argument:
– Effort is a public good to the members
– Public goods are underprovided
– Therefore, effort being underprovided, labor
productivity would be low.
• In fact, labor productivity is no lower, and often
higher, than in comparable capitalist firms.
• Worker-owned plywood companies were formed in
Washington, Oregon, and nearby largely in the period
• About 1950 they produced about 25% of output, all of
which came from the Pacific Northwest region.
• Many were quite successful, but some of the most
successful were sold to private-enterprise corporations at
• Others have gone out of business due to the “successor
• Membership was based on the purchase of a share. In some
cases I have observed, the price of a share would
approximate the price of a modest house.
• The Board of Directors often must approve the purchase of
• One man, one share, one vote.
• The members elect the board of directors who hire the
• Some nonmembers might be employed (in a few cases a
majority) but shareholders have priority when there are job
openings, and nonmembers would be laid off first.
• Dividends on shares were limited, and usually not paid at
all. Some associations prohibited them.
• Hourly wages were equal.
• Jobs could be quite different, some dangerous and
physically demanding. Job rotation was common.
• Managers had great power to shift job assignments, and
discretion generally, though they sometimes complained
that the members took too much interest in management.
• They also complained that the members were too short-
term in their interests.
• The plywood coops spent less on supervision than
comparable firms -- a fraction as much.
• “When one cooperative was converted to a privately
owned mill, the number of supervisors and foremen was
• Labor productivity was no less, and some studies
indicated that it was greater, than in comparable capitalist
• The cooperative firms responded differently to changing
circumstances than comparable capitalist firms.
• In the face of a decline of demand, they reduced wage
levels but were less likely to lay off and laid off fewer
• In the other firms, the main response is a reduction of labor
hours, employment, and production.
• This is consistent with the criticism that LME’s would
“misallocate labor” but also suggests that, in a system
dominated by LME’s, recessions would be less contagious.
• (The evidence is also consistent with Marxist theory of
• Mondragon Cooperative Corporation is a group of worker
cooperatives in the Basque country of Spain.
• A leading figure was Catholic Father Arizmendi, who
founded a polytechnic school in 1943.
• The first company was FAGOR, 1956, which originally
manufactured kitchen equipment.
• (My wife has a FAGOR pressure-cooker.)
• Caja Laboral Popolar, a mutual bank, was founded in
• Its credit supported the start-up of new cooperatives and
the conversion of some bankrupt firms to coops.
• The group of enterprises is highly interdependent --
finances are through the Caja Laboral, they give discounts
to one another, and when one coop fails or has to lay off
labor, they may be shifted to other coops.
• The group constitutes the largest corporation in the Basque
country and the seventh largest in Spain.
• Their retail store chain is the largest in Spain.
• Governance is federal, with an elective Cooperative
Congress for the whole group but also autonomous
workers’ councils in member coops.
• Among the requirements for membership in the group, and
access to finance through the Caja Laboral, is the
requirement that a proportion of net revenues be
• Members are required to buy in, a fairly substantial
investment, though it can be financed partly by a loan from
• As a result, the group has grown consistently and rapidly.
• The group operates a university, and about half of the
graduates work in the Mondragon companies.
• The group employs about 100,000+, though in 2007 only
about 1/3 were members.
• Most non-members are at locations outside the Basque
region. Globalization has required expansion, including
other countries were membership is considered
• Within Spain, many are members of the retail chain,
• It is expected that the membership will be increased to
about 70%, with Eroski employees offered buy-ins to
• Mondragon is widely seen as a model to be
emulated by worker coops elsewhere.
• There have been other such initiatives in Spain,
with some success, and in Canada (which has a
long cooperative tradition of its own.)
• Could it be done in the US?
• The Legacoop, in Northern Italy, includes both worker and
consumer coops, and mutual banks.
• Its regional success has been comparable to that of
Mondragon, mainly in Romagna-Emilia.
• Labor productivity studies parallel those for the plywood
• So do studies in several other countries.
• In France, worker coops are required by law to reinvest
some part of their net revenues, and cannot distribute it on
• There is evidence that French coops are pretty long-lasting.
• It seems clear that capital funding remains a problem for
worker coops -- a soluble problem (as in Mondragon) but a
• Reinvestment is useful but could hardly lead to an efficient
allocation of investment.
• But cooperators are very reluctant to raise capital on
capital markets, as they believe they would lose control to
shareholders, and even with loans they might lose control
• In 1977 I made a proposal that I thought might enable
worker coops to participate in capital markets, but it is
• The evidence from worker cooperation seems to establish
that democratic worker participation in the governance of
firms is feasible.
• There is no sacrifice of labor productivity. Indeed, on the
whole, it may be advanced.
• Investment decisions might be distorted, but this seems to
vary greatly depending on the details and form of the
cooperative and the cooperative sector -- so probably
• Could a society based on the ideals of democratic
cooperation be established?
• So what if a group wanted to form a coop this year?
– Here is a proposal I made at a conference at Mondragon
– Charter 2 corporations, one conventional for-profit and
– The mission of the not-for-profit would be to promote
employment and incomes through financing and
encouraging worker cooperatives.
– The founders would lend whatever capital they can
raise to the nonprofit.
– The nonprofit would buy all the stock of the for-profit.
– The nonprofit would delegate governance of the for-
profit to the elected representatives of the for-profit, on
the condition that the shares pay dividends proportional
to the average compensation of employees. For some
principles to determine the proportion, see
– The dividends are used to repay the loans to the coop
and subsequently to finance further expansion of coops.
• Recall, from third week, the contrast of the Marxist and
Schumpeterian theories of the self-destruction of
• Neither quite corresponds to the facts of the twentieth
• Here is a synthesis:
– Capitalism leads to recurring, increasingly severe
– To mitigate them, governments introduce, in the
interest of the capitalist class, noncapitalist elements
that gradually abolish capitalism.
Self-Abolition of Capitalism
• Following the crisis of the 1930’s, Keynesian demand
management, social security and safety nets were
• In the 1980’s, temporary nationalization of banks.
• In 2008, conversion of corporations to a sort of hybrid
• And so on.
• Free-market conservatives have argued since the 1930’s
that “liberal” measures just set the stage for new crises and
gradually abolish capitalism. What if they are right? The
choice seems to be to abolish capitalism gradually or let it
collapse now. Has capitalism a long-term future?