This article was published in the Autumn 1995 issue of Formulations
by the Free Nation Foundation
Good and Bad Collective Action
Can We Nourish One and Squelch the Other?
by Roderick T. Long
This paper will be presented at our 14 October 1995 Forum.

(to table of contents of FNF archives)   (to start of essay)

A Problem for Libertarians
Love, Hate, and Greed
Long-term vs. Short-term Greed
Reality Check
Reasons for Optimism
Who Needs Collective Action?
Culture and Collective Action

  (to outline)   (to top of page)

A Problem for Libertarians

How easy, or difficult, would collective action be in a free nation?

This is a question to which we, as libertarians, might seem committed to giving inconsistent answers. When the collective action in question is something good or desirable, we're confident that market incentives and natural human sympathies would unite to bring the collective action about without the need for coercive coordination from government. But when it comes to harmful or unpleasant collective action, this, we are sure, can flourish only with the help of state intervention, and will quickly wither and die when exposed to the light of freedom and economic rationality.

Consider the problem of racial and sexual discrimination. Discriminatory hiring practices represent a form of collective action, in that a pattern of discrimination against the same groups occurs throughout society. (If discrimination didn't follow a common pattern, it would be far less problematic. That is, if it were a purely random matter which groups were discriminated against by any one employer, then those who experienced discrimination from a given employer could be sure of finding plenty of other employers who lacked that particular prejudice. The prejudice might still be a vice, to be sure, but it would at least be a harmless vice. It's only when there's a consistent and widespread prejudice throughout society against certain groups that members of those groups find themselves systematically disadvantaged across the board. This result is what makes discrimination so especially objectionable.)

Discrimination in hiring is a problem we like to think would be solved by the free market. Firms that choose their employees on the basis of race or gender, instead of on the basis of merit, will end up with a less capable workforce, and so the firm's overall performance will suffer, thus exposing it to the risk of being edged aside by their competitors. Thus, rational firms, in their pursuit of the Bottom Line, will have to abandon their discriminatory practices on pain of losing out to the competition. Thus homo śconomicus comes to our rescue: racism and sexism are simply too expensive. They represent costly luxuries in which a competitive firm cannot afford to indulge.

This argument assumes that economic self-interest is likely to be a more powerful motive than such purely emotional motives as racial and sexual prejudice. But libertarians do not always make this assumption. When it comes to the provision of public goods, we suddenly start to heap scorn on the narrow homo śconomicus conception of human motivation that served us in such good stead in the prejudice case. Now we want to insist that economic self-interest is not the only human motive, that incentives such as conscience and solidarity can override the quest for profit. The relentless concern for the Bottom Line that turned up so conveniently to impede harmful collective action, now just as conveniently drops out so as not to impede beneficial collective action. What entitles us to this double standard?

  (to outline)   (to top of page)

Love, Hate, and Greed

All human motivations can be divided into three categories, which I shall label, rather simplistically, love, hate, and greed. Under love I rank all those motives that have as their end the satisfaction of the legitimate interests of other people. Under hate I rank all those motives that have as their end the frustration of those interests. And under greed I rank all those motives whose ends make no essential reference to the interests of others one way or another. (A person acting from greed may harm or benefit others, but only insofar as doing so happens, under the circumstances, to advance his ends. Greed as such is indifferent to the interests of others.)

The first thing we should recognize is that motives of all three varieties are available in plentiful supply. Any account of human nature that emphasizes just one of these motives at the expense of the other two can safely be dismissed as unrealistic.

Now we can see that the libertarian responses we gave to the public-goods and prejudice problems seem to assume that greed is stronger than hate but weaker than love. When the racist employer hires the minorities he despises because it's good for business, greed is conquering hate. When the public-spirited citizen contributes to a public good out of a sense of moral duty or communal solidarity, although he could get away with free riding, love is conquering greed.

It would be delightful, of course, if greed could be counted on to be strong in its conflicts with hate and weak in its conflicts with love. But we know, all too well, that motives of hate can often conquer motives of love; so there is no guarantee that love is always strong and hate is always weak. Thus it's not implausible that hate should often be strong enough to conquer beneficial greed, or that love should often be too weak to prevail against harmful greed.

  (to outline)   (to top of page)

Long-term vs. Short-term Greed

A similar tension can be found in libertarian discussions of conflicts between different kinds of greed. Consider the many cases in which it's in my long-term interest to acquire a reputation as a cooperator, while it's in my short-term interest to renege on cooperation just this once. Which are people in a free society more likely to do?

When the cooperation is a beneficial one, we rush to say that long-term greed will win out. Citing such works as Robert Axelrod's The Evolution of Cooperation, we point out that cooperators, by developing a reliable reputation, will attract a cluster of like-minded cooperators to them, whereas habitual defectors will be shunned and excluded from the benefits of cooperation, so that both market competition and natural selection will tend to make cooperation prevail as a strategy. Actors in the market will realize that the benefits of keeping to a consistent policy of cooperation outweigh the meretricious short-term gains of opportunistic defection.

But sometimes cooperation is not so nice, and then we tend to have a different attitude. Consider the standard libertarian response to the problem of cartels. In an unregulated free market, what would prevent profit-minded firms from joining together and agreeing to keep prices high, or wages low? We usually answer that once the cartel is in place, it's in the interest of any individual member to break the agreement by selling at a slightly lower price or hiring at a slightly higher wage, so as to win all the other members' business for oneself. Soon, we like to predict, all the members will be tempted into trying the same strategy, and the cartel will collapse.

But what has now become of the idea that rational individuals will choose to maintain a system of cooperation rather than defect for the sake of immediate gain? Axelrod has been thrown to the winds! Short-term greed, so fragile a hindrance to beneficial cooperation, now proves itself a formidable bulwark against harmful cooperation. Long-term greed, on the other hand, has dwindled from its former glory as guardian angel of cooperation, and now is nowhere to be seen.

This now-you-see-it-now-you-don't act proves particularly embarrassing for libertarians trying to defend market anarchism. What ensures that, in the absence of government, private protection agencies will choose to resolve their differences through arbitration rather than violent conflict? Long-term greed, which recognizes that the value of maintaining a system cooperation outweighs the value lost by submitting to arbitration. But what ensures that these protection agencies won't merge into a giant cartel, thus in effect bringing back government? Short-term greed, which undermines cartel agreements in the usual way. The balance of motivational power between long-term and short-term greed keeps swinging back and forth, as needed by our libertarian arguments.

This is cause for worry.

  (to outline)   (to top of page)

Reality Check

The problems I've been pointing to should make us uncomfortable. But they shouldn't necessarily drive us to despair. Perhaps we can make ourselves feel a little better by noticing that all the mechanisms we like to trumpet have actually proven successful in the real world in many, many cases.

Consider first the case of prejudice. It's no coincidence that there were Jim Crow laws in the pre-civil-rights South. White racists were unwilling to rely on voluntary compliance alone to keep blacks "in their place." And this reluctance on their part was a shrewd one. The famous segregated buses in Montgomery, Alabama, were segregated by law, not by the choice of the bus company. In fact, the bus company had petitioned, unsuccessfully, to get the law repealed — not out of love but out of greed. So we're quite right in thinking that racism can be undermined by a concern for the Bottom Line (though it would be naïve to assume that it must always be so undermined; people do care about things other than money, and some of those things are pretty repugnant).

It's also true, of course, that people voluntarily contribute to good causes all the time. The amount of money given to charity every year in this country (over and above taxes!) is staggering.

Similar remarks apply to the issue of long-term vs. short-term greed. Beneficial collective action occurs all the time without coordination by government; are cooperative impulses are the product of evolution, and are reinforced by our social environment. To pick just one example mentioned by Axelrod, soldiers on opposite sides of World War I trench warfare found it in their mutual interest to coordinate their firing patterns in such a way that each side would know when and where the other was going to fire and so could avoid injury. Score one for Axelrod. On the other hand, history is full of cartels collapsing because of members' breaking the agreement in order to reap the benefits of underselling; one such defection (by Kuwait against oil partner Iraq) triggered the recent Gulf War. Score one against Axelrod.

These examples should serve to reassure us that our analyses of collective action problems aren't simply drawn from some fantasy world unconnected to reality. But can we say anything more than this?

I think perhaps we can.

  (to outline)   (to top of page)

Reasons for Optimism

We would have stronger reasons for optimism if we had some reason to think that the motives for harmful cooperation had some weakness, some fatal flaw, that the motives for beneficial cooperation did not share. I think there is at least one such weakness.

Notice that the motives for harmful cooperation are motives for selective cooperation. The white racist who cooperates with other white racists in discriminating against blacks is not taking a cooperative attitude toward the blacks themselves; likewise, those who cooperate to form a cartel are colluding to engage in decidedly non-cooperative behavior toward their customers. The same holds true for protection agencies in a state of nature that combine to form an oppressive government. In all three cases, the cooperation in question is cooperation for mutual advantage within a select group, and is directed against the advantage of those excluded from the group. Such cooperative ventures are easier to undermine when there is free competition, because they create a large group of excluded people who have an interest in seeing that cooperation end, and this group constitutes an attractive market for any entrepreneur interested in defying the cooperative venture.

To be sure, pressure within a selectively cooperative venture of the kind I've described may be strong enough to discourage defections. The racist, tempted by profit to hire the qualified black over the unqualified white, may think again when he realizes he will be subject to severe social sanctions from his fellow racists within the community. The pull of the Bottom Line can be quite limited in the face of social ostracism by one's peers.

But that's precisely why I stress the importance of free competition. The beneficent power of greed in overcoming harmful cooperative ventures lies not so much in its ability to undermine the venture from within, as in its ability to attract rival cooperative ventures to outcompete the bad ones. The white racist who has lived all his life in Kluxville may prefer social conformity to profit, but if the resulting low wages for blacks in the Kluxville area serve as a cheap-labor magnet motivating Amalgamated Widgets to open a new plant in Kluxville, the folks who run Amalgamated Widgets may not care that much if the whites in Kluxville shun them; they already have their own peer group, after all.

The ease with which the greed of outsiders can defeat the hate of the exclusive group (or, switching to the cartel situation, the ease with which the short-term greed of outsiders can defeat the long-term greed of the exclusive group) depends on the degree of competition. If regulations make it extremely difficult to start new ventures or expand old ones, then there will be a smaller number of long-established players, insulated from competition and so free to try their hand at harmful cooperation. (Thus government regulation may be described as subsidizing racism and cartelization. That's why large corporations in America during the "Progressive Era," and racists in South Africa at the beginning of apartheid, were such enthusiastic fans of government regulation.) The easier it is for a new venture to start up, the easier it is for harmful cooperative ventures to be undermined from without. Assuming free competition is present, it is the selectivity of harmful cooperation that spells its death knell.

Beneficial cooperation is not selective in the same way. That is not to say that a virtuous cooperator cooperates with everyone equally. Any cooperative venture — be it a family, a business, or a political movement — is focusing more on the advantage of its participants than on the advantage of outsiders. But that kind of preferential concern is not the same thing as a concerted opposition to the welfare of outsiders. What creates trouble for the bad cooperative ventures is that they create an aggrieved, excluded class which forms the natural market for a competitor to enter the field. Mere preferential concern alone does not do that.

It might be objected that at least one beneficial cooperative venture — the private protection industry in a market anarchist society — creates at least one excluded class — criminals. Doesn't this create an incentive for a competitor to enter the field and offer criminals the wherewithal to fight back against the protection agencies?

It surely does. Hence organized crime might exist in a libertarian society. After all, libertarians are fond of pointing out that governments in effect subsidize organized crime by prohibiting, and thus creating an attractive black market for, such victimless crimes as prostitution and drugs. But a libertarian legal system, whether minarchic or anarchic, would at least prohibit victimful crimes (i.e., crimes that do have victims) like murder, theft, assault, rape, arson, fraud, and the like, and so, by the same reasoning, would create a black market for these crimes.

Still, cooperative ventures against victimful crimes are likely to be more successful than ones against victimless crimes, precisely because the former has a crucial source of support — namely, the victims (and potential victims) — that the latter lacks.

This brings to mind a related point that was first brought to my attention by Bryan Caplan. Some boycotts are self-enforcing while others are not. For example, if I have a policy of refusing to do business with anyone who doesn't belong to my religion, this policy will clash with my financial incentives. The financial incentives may still lose out, of course; but then again they may not. On the other hand, if I have a policy of refusing to do business with people who cheat their customers, my financial incentives are likely to reinforce this policy. Choosing criminals as one's target market is risky precisely because people who make a profession out of non-cooperation cannot be relied on to cooperate with you either. (That's one reason that the most successful criminal organizations have been ones whose members shared some ethnic, religious, political, or family connections, making them less likely to defect with each other than with outsiders.)

  (to outline)   (to top of page)

Who Needs Collective Action?

I should also point out that the need for beneficial collective action may be overstated. As I have argued elsewhere ("Funding Public Goods: Six Solutions," in Formulations, Vol. II, No. 1 (Autumn 1994)), collective action on the basis of love or long-term greed is only one way to provide public goods. Another way is to privatize the public good, either absolutely (i.e., by figuring out some way to exclude non-contributors) or else by packaging it with a private good, and using the revenue from the private good to fund the public good (i.e., using advertising to pay for radio and TV broadcasts, or using harbor fees to fund lighthouses). So the fact that beneficial collective action is not 100% reliable is no reason for despair, given that the same ends can often be achieved through non-collective means.

  (to outline)   (to top of page)

Culture and Collective Action

Let me close by considering some of the ways in which cultural factors can influence the success or failure of collective action.

There are two reasons collective action can fail. One reason, the reason we've been discussing so far, is motivational. Collective action can fail because not enough people want to participate in it.

But the other reason is informational. Suppose everyone in Shangri-la wants to go on a general strike to protest the actions of the government. There's no motivational problem here; everyone wants the same thing. But there's an informational problem: when should the strike begin? If only a few people start on their own, they'll simply be punished and nothing will be achieved. As in many cases, the acts of resistance must be simultaneous in order to be effective.

This is a coordination problem. The key to solving such problems is what rational choice theorists call salience.

What is salience? The classic example goes like this. Suppose you and a friend intend to meet in New York City on a specific date. Unfortunately, neither of you will be able to contact the other ahead of time to arrange a time and place to meet. So you have to try to find your friend (and your friend has to try to find you) with no more specific information than the city and the day.

What should you do? Well, you should go wherever you think you friend would go. But your friend is trying to figure out where you would go ....

The answer most people give — which in effect makes it the right answer — is that you should go to Grand Central Station at noon. In New York, Grand Central Station is an "obvious" meeting place, and noon an "obvious" meeting time. That place and that time stand out from their competitors. They have salience.

Salience is likewise what the Shangri-la strikers need. If there is a tradition in their culture of going on strike on a certain date, that is the date to pick. In the absence of any such tradition, something else is need to provide the salience. That is one function of a leader. If there is some one person whom the strikers all respect, that person can make a particular date salient by saying, "let's strike then!"

One might also see salience as a way for people to get themselves from an unproductive cooperative venture into a productive one. After all, resistance to an oppressive government is an instance of collective action, but so is the existence of that government itself. I don't just mean that the rulers in the government are cooperating with each other; in some sense, the ruled have to be cooperating too in order for the government to be effective. Rulers have power only so long as people obey them. And why do people obey them? Partly because they think it's their duty to do so, or else because they think they can benefit from government power; to that extent, overthrowing a tyrannical government runs up against a motivational problem. But partly also because they're afraid to be the only person resisting the government. Even if everyone hates the government, there's still the problem of knowing when and how to resist. In that case, salience can help people escape a trap of their own making. To switch from obedience to resistance is to switch from one mode of collective action to another; and, like switching from driving on the right to driving on the left, people are going to get run over unless the switch is made en masse rather than one person at a time.

To the extent that prospective cooperators share a common cultural background, it will be easier for them to overcome both the motivational and the informational obstacles to cooperation. Motivationally, people from the same culture are more likely to have similar values and a feeling of solidarity, and so will be more willing to cooperate with one another. Informa-tionally, it will be easier for people from the same culture to find salient points to build coordination on, since they share either a common tradition or a common set of leaders or both.

Consider a mediæval case of collective action. In the Middle Ages, the Catholic Church promulgated the Peace of God (forbidding warfare during certain months of the year) and the Truce of God (forbidding warfare during certain days of the week). These restrictions on warfare were fairly widely observed, with extremely beneficial results to all parties concerned, since adherence to these rules prevented warfare from becoming all-consuming, and allowed the usual business of life — commerce, agriculture, and so on — to continue relatively undisturbed. But this beneficial collective action was possible only because the warring parties shared a common allegiance to the Catholic Church. Their religious faith gave them a motivation to obey the Church, and the Church's authority made the particular provisions of the Peace and Truce salient. By contrast, when Christians fought Muslims there were no such constraints, because the combatants lacked a shared cultural basis to support anything like a Peace of God or Truce of God.

Having a common culture makes bad collective action easier too, however. As I've pointed out before ("Can We Escape the Ruling Class?," in Formulations, Vol. II. No. 1 (Autumn 1994) and "Religious Influence on Political Structure: Lessons from the Past, Prospects for the Future," in Formulations, Vol. II, No. 3 (Spring 1995)), adherence to a common religion on the part of the common people was a large part of what held the ruling classes of ancient and mediæval societies in power, since this religion taught that those in power ruled by divine right. In a more pluralistic society, it would be much harder for any one group to claim a divine mandate, and so such ruling cliques should be easier to oust.

What, then, are the cultural prospects today for collective action, good and bad? That depends on whether we are moving toward or away from cultural unity, and that's not an easy thing to tell. Within each society we see a great deal of pluralization and splintering going on. But we also see a great deal of homogenization going on between and among different societies. So it's difficult to say whether collective action in general is going to become easier or more difficult.

The problem of collective action in a libertarian society is one I don't claim to have worked out fully by any means. I offer this discussion in the hope that it will stimulate debate. For debate, as John Stuart Mill teaches us, is a discovery process that assists us in our search for answers. D

(to table of contents of FNF archives)   (to outline)   (to top of page)