In our last week's postings on the French riots precipitated by the new employment law, now in force, that allows employers to fire, without cause, during the first two years of employment employees under the age of 26, Gary Becker and I offered some possible explanations for the riots that did not depend on any notion of a distinctive French political or social culture. But it is difficult to resist attributing some causal efficacy to cultural factors. The new law permits, though within a very limited scope, employment at will, the dominant employment contract in the private sector in the United States--and public opinion polls indicate that the French are more hostile toward capitalism than virtually any other nation; a recent poll finds that only 36 percent of the French have a favorable opinion of capitalism, compared to 71 percent of Americans. The French are at once highly statist, which is related to the rejection of capitalism, and highly prone to riots and work stoppages, which may also be, as I will suggest. They are unusually xenophobic, as indicated by their efforts to prevent the incorporation of English words into the French language, their resistance to foreign acquisition of French companies, and their emphatically independent line (going back to De Gaulle) in foreign relations and military matters. The French are, in short, culturally distinct.
Can an economic account of cultural difference be given? If so, the starting points would be habit, on which Becker has written, and the economics of organizations, networks, and language, on all of which I have done some work. Much behavior is habitual, and this can be given an economic meaning. Behavior is habitual when it is done without conscious thought, or more precisely with limited thought. (More technically, behavior is habitual when cost and beneÔ¨Åt are time-dependent and cost is negatively related to time and beneÔ¨Åt positively related to it. See Gary S. Becker, "Habits, Addictions, and Traditions," 45 Kyklos 327, 336 (1992), and notice that the obverse case‚Äîcost positively, beneÔ¨Åt negatively, related to time‚Äîis that of boredom.) One is conscious that one is brushing one'‚Äôs teeth, but the amount of thinking that is involved is very slight; one does't think about which tooth to brush next, how long to brush, how much toothpaste to put on the brush, and so forth. So transforming behavior from deliberated to habitual, usually by repetition, economizes on cognitive effort, and is thus cost-effective. The cost of making it habitual is a sunk cost, moreover, which means that once behavior is habitual it is cheap to continue with it. This discourages change, since new costs would have to be incurred.
Culture is similar. The way in which people speak (including pronunciation, grammar, syntax, and vocabulary), gesture, hold their bodies, grimace, use knife and fork, greet one another, behave toward members of the opposite sex, and otherwise conduct themselves in their basic social interactions is to a high degree habitual, most of that behavioral repetoire having been learned and mastered in early childhood--when learning costs (notably of language) are low. To change one's cultural identity as an adult requires incurring heavy time costs, often with limited results--a foreigner is unlikely ever to lose all traces of foreignness. This is one reason why there are many different languages, even though it would be more efficient (costs of change aside) if everybody in the world spoke the same language.
Beliefs also have an important habitual element. If one is brought up to believe that being American or French is very special and that foreign attitudes and behaviors should be held at bay, the adherence to those beliefs is cheap. Once you believe something, you will be reluctant to incur the cost of changing the belief unless assailed by some doubt that you cannot easily resist; and people do not like being in a state of doubt. Most people cannot give a good account or defense of their beliefs; they believe something because they have always believed it, not because they have a good, conscious reason to believe it. Once a belief system is entrenched, it is likely to persist indefinitely until a very rude reality check causes the costs of continued adherence to the system to exceed the costs of change. That has not yet happened in France, a rich country--surprisingly so, given its high unemployment rate and overregulation. GDP per capita is the 21st highest in the world, and though significantly lower than that of the United States is comparable to that of Germany and the United Kingdom, and higher than Sweden's. In addition, France has the world third-highest life expectancy and its workers have a great deal of leisure. The country can afford high unemployment and a short work week for those who are employed because it has the most productive workforce in the world, though in part this is an artifact of high unemployment--the unproductive workers aren't employed. The high standard of living of the French, which actually understates the quality of French life because of their leisure and life expectancy, discourages the French from reexamining their political beliefs, since such reexamination would be costly, whereas continued adherence to their beliefs is costless in the sense of requiring no mental energy and not preventing the enjoyment of a good life by most people. A weak work ethic weakens resistance to the inconvenience caused by riots and work stoppages.
There is an interesting literature on organizational culture--the beliefs, norms, customary practices, methods, etc. that are found in, and vary among, particular organizations--that bears on our subject. The literature emphasizes the difficulty of changing an organization's culture. A famous study of our failure in Vietnam, by Robert Komer (Bureaucracy Does Its Thing: Institutional Constraints on U.S.-GVN Performance in Vietnam (R‚Äì967‚ÄìARPA, Aug. 1972), and a similar recent study by John Nagl (Learning to Eat Soup with a Knife: Counterinsurgency Lessons from Malaya and Vietnam (paperback ed. 2005)), document the inability of our government to change the organizational culture of our armed forces, even though it was plain to many people in government that that culture was poorly suited to the conditions of the Vietnam war. Business failures frequently stem from the same cultural inertia, even though business receives from the market strong signals that its culture is maladapted.
The explanation is in part the sunk-costs problem that I have emphasized, but also the "network" character of an organization's culture, which makes it necessary, in order to change the culture in a constructive way, to change a great many things at once, and some of the things may be very difficult to change. For example, when the Detroit auto companies began losing market share to Japanese manufacturers, they studied Japanese methods and decided to adopt one of them, the "quality circle" approach whereby workers were encouraged to take an active role in suggesting productivity enhancements. Transposed to the United States, the approach failed because workers realized that improvements in productivity might endanger their job. This wasn't a problem in Japan because workers in the Japanese automobile industry had de facto lifetime tenure, which our workers did not.
In Vietnam the obstacles to our armed forces' changing their organizational culture was that the culture was optimized to a continuing threat, namely that of a conventional war in Europe. Secretary of Defense Rumsfeld has had a somewhat easier job of altering the armed forces' culture to make it more responsive to what is being called "postmodern warfare," illustrated by the struggle against global terrorism and the insurgency in Iraq, because the threat of a conventional war has receded. In addition, in seeking to transform the armed forces, Rumsfeld is aided by the fact that technological innovation, the key to his concept of transformation, is an element of the traditional U.S. military culture.
So culture is habit writ large and is difficult to change because habits are difficult to change. And changing national cultures, like changing social norms and customs (themselves a part of culture), is particularly difficult, because of the lack of centralized direction. In principle, the culture of a nation's military can be changed by an order of the commander in chief, though in practice such an order is likely to be foiled by passive resistance within the organization. But a national culture has no hierarchy, unless the nation is totalitarian. To change a nation's language or any other deep-seated feature of its culture requires coordinated action without a coordinator. Imagine a country whose leaders thought it would be more efficient for drivers to drive on the right-hand side of the road rather than the left but decided to leave the change to the spontaneous decision of the drivers.
The dramatic though not complete changes in the German and Japanese national cultures that occurred in the wake of World War II were facilitated by the fact that in each case the war had smashed much of the existing national culture as well as demonstrated its dysfunctional character. That has yet to happen in France.
Posner emphasizes two important reasons, habit and coordination costs, for why cultures, including those of nations and companies, often change very slowly. I agree with his arguments on why habitual behavior and coordination-network effects are important, and do not have much to add directly to his discussion. Instead, I will try to explain the seeming contradiction that many aspects of national cultures change drastically and often surprisingly quickly. In a nutshell what I claim is that major economic and technological changes frequently trump culture in the sense that they induce enormous changes not only in behavior but also in beliefs.
A clear illustration of this is the huge effects of technological change and economic development on behavior and beliefs regarding many aspects of the family. Attitudes and behavior regarding family size, marriage and divorce, care of elderly parents, premarital sex, men and women living together and having children without being married, and gays and lesbians have all undergone profound changes during the past 50 years. Invariably, when countries with very different cultures experienced significant economic growth, women's education increased greatly, and the number of children in a typical family plummeted from three or more to often much less than two. Divorce rates often exploded, men and women married later, and living together without marriage became more common.
Ireland is an excellent example since not long ago Irish family patterns were the object of study by demographers only because they were so different. These patterns involved late ages at marriage, high birth rates, no divorce, and married women who spent their time mainly caring for children and their husbands. Enshrined in the Irish Constitution of the 1930's is the hope that married women would not work but instead they would be home taking care of their families.
All aspects of Irish family behavior changed radically during the past two decades: the typical family now has only about two children, divorce was legalized and is growing rapidly despite the Catholic Church's opposition, and the labor force participation of married women is becoming like that in other parts of Western Europe. The rapid economic growth Ireland experienced during the past couple of decades had a revolutionary impact on the incentives of parents to have many children, on attitudes about whether married women should work, and on whether married couples were obligated to remain together throughout their lives. What is fascinating about the Irish example is that these and other changes in family patterns of behavior occurred while Ireland remained a highly devout nation, with the highest rates of church attendance and other measures of religious belief in the Western world.
Similar changes have also occurred in very different cultures. Japan, South Korea, Taiwan, Hong Kong, and other Asian countries had rapid declines in birth rates from very high to very low levels as they experienced extraordinary economic growth during the second half of the twentieth century. Married children are much less likely to live with parents than before their economic development, divorce rates are rising, and married women have become active in the labor force. Birth rates have declined and other aspects of family behavior have changed also in the primarily Muslim country of Malaysia, although these changes are generally smaller than in neighboring countries that experienced economic development.
Even the "unique" culture of the French changed enormously as it experienced rapid development after World War II. French birth rates also fell below replacement levels, although they are higher than in other parts of Western Europe (partly due to generous child subsidies in France), married women work in large numbers, and divorce rates have risen a lot. Nor are the radical changes in France restricted to the family. A leading French businessman spoke in English before an international audience recently because he said English is the language of commerce. President Chirac walked out before he began to protest his speaking in English. The command of English has been growing rapidly among French youth, and many French companies conduct their board meetings and other internal discussions mainly, sometimes entirely, in English. French McDonald's restaurants have been attacked, but they are highly profitable because many French young people go there. English words are growing rapidly in spoken and even written French, despite continual protests.
Yet, I am not suggesting full economic determinism, a la Marx. While economic progress and technological change do have enormous effects on "culture", some important national institutions, attitudes, and behavior change much more slowly. Among other examples, Posner mentions language, attitudes toward globalization, whether one drives on the left or right sides of a street, and units of measurement. Even various dimensions of family life change much slower in some cultures than in others. For example, while divorce rates have risen at rapid rates in Japan and South Korea, they are still much lower than those in the West, and the labor force participation of married women is also much lower.
Change is easier in those parts of culture that are due mainly to habit. For habits respond as economic circumstances, technologies, and incentives change. At first, habitual behavior is usually slow to change since past behavior exercises enormous influences over current behavior that is "habitual". But the initially slow changes induce further and more rapid changes in later behavior, so that the cumulative change may eventually be quite big. Smoking is habitual, even addictive, but that did not prevent large declines in smoking rates in all developed countries since the early 1970"s after the evidence on its connection with cancer and heart disease became evident. In the past it was expected that women would stay home after marriage and have large families. These expectations changed greatly and rather rapidly as women"s education and their job opportunities improved, and as the economic advantages of investing more in the education of children became apparent.
Change is harder in behavior that requires detailed coordination among different citizens. This is why measurement systems, basic street driving rules, the official language, and some other aspects of a country's institutions change very slowly, although languages get infiltrated by foreign and new words, and greater emphasis is placed on languages that are more important for communication with other countries. Despite French protests, English has become not only the international language of commerce, but also of advanced education instruction, and of communication through the Internet.
To relate this to last week‚Äôs topic, how much of the opposition in France to the proposed youth labor law can be attributed to a special French culture? The propensity to riot is higher in France than say Germany, but strong and effective opposition to making discharge of employees easier is found also in German, Italy, and Portugal. Nor is the opposition restricted to Western Europe. So far the free market oriented Indian Prime Minister, Dr. Manmohan Singh, has not been able to make much of a dent in India's similar employment system, and the same difficulties in discharging employees are found in most of Latin America.
No, the conflict over the proposed youth labor law is not particularly "French", but basically comes from the opposition of labor market insiders with good jobs who do not want to give up their privileges. This is the same source of opposition to labor market reforms in France, Germany, Italy, India, Argentina, and other countries with widely different cultures. It is more interesting to ask why Anglo-Saxon countries like Great Britain, Australia, and New Zealand have moved toward flexible labor markets in response to the effects of globalization, whereas these other nations have not? I do not have the answer, but before we attribute this to a unique Anglo-Saxon culture, let us recognize that China has also introduced flexible hiring and firing practices as it moved rapidly to a market economy. China obviously has a radically different history and "culture" than Anglo-Saxon nations.