Why Do They Not Comply with the Law? Illegality and Semi-Legality among Rural-Urban Migrant Entrepreneurs in Beijing

Abstract

Based on in-depth interview materials, this article examines why most rural-urban migrant entrepreneurs in Beijing do not fully comply with a discriminatory license requirement, and in particular, why they prefer license-renting from the locals. This article suggests that the law's lack of legitimacy adds weight to instrumental considerations. But more important, this license-renting practice seems to be reinforced and sustained institutionally by local businesses, law enforcement officers, and the local authorities, because their interests are inextricably intertwined with it. The whole situation constitutes a general equilibrium through which various interests are balanced. This case study thus paints a far more complicated picture of the law's impact on people's behavior than usually assumed. Instrumental concerns, or coercive action and sanctions alone, do not adequately explain people's interaction with the law in a “lawless” circumstance; a whole range of instrumental concerns must be considered, and they, together with sanctions, must be understood in the context of a larger institutional environment in which the interactions of various players unfold.

After more than two decades of legal reforms, China has established a series of legal regulations and institutions from top to bottom that basically mimic what Western countries already have. Institutions and regulations alone, however, do not guarantee the rule of law (Weber 1958). Observers soon noticed a prevalent noncompliance with these legal regulations during the reform period. Of course, exact and total compliance never truly occurs, but compared to developed countries, the situation in China seems to be quite outrageous. To explain the widespread noncompliance, some have drawn attention to the general institutional and cultural obstacles that hinder people in China from accepting the law (Epstein 1994:19; Potter 1994). Others argue that the institutional environments of transition in China have resulted in corruption, rent-seeking, and various forms of patron-client relations (Young 1989; Rose-Ackerman 1999; Wank 1999; Wedeman 2003). While these studies portray a general picture of the situation, the ways in which the populace views, adapts, and reacts to these regulations in the real world are inadequately addressed.

Taking an approach at the grassroots level, this article examines the behavior of a group of people—rural-urban migrant entrepreneurs—conducting business in Beijing, China, who face serious discrimination imposed by licensing legislation. Compared to the locals, it is far more difficult for the migrants to obtain the business licenses they need to conduct business lawfully as individual business operators (getihu). Faced with this barrier erected by the law, some migrants obtain licenses and others do not. A considerable proportion of these migrants rent licenses from the locals; technically, this is illegal but is countenanced as a form of collusion among migrants, local businesses, and local officials.

In explaining the migrants' choices and, in particular, why they prefer renting licenses in Beijing, we can hardly draw on resources from the current studies on legal compliance because most of these have been conducted in liberal democratic societies, where legitimacy of the law is often assumed (Tyler 1990). As I shall illustrate, however, the legitimacy of the licensing legislation barely exists among those migrant entrepreneurs. The key question, then, is: in such a situation, what are the migrants' considerations when dealing with a system that might be labelled as “lawless regulations”? Comparative studies in post-Communist countries indicate that compliance by the people is largely due to incentives and sanctions (Feige 1997). Yet to what extent is this observation applicable in the China case?

Based on empirical data from a variety of sources,¹ I suggest in this article that, in the China case, the law's lack of legitimacy does add weight to instrumental considerations. To migrants, the authorities and the law are not worthy of respect, and it is not in their practical self-interest to obey them. By pondering on the illegal, legal, and semi-legal options for license-renting, migrants have found semi-legality the best choice. More important, the semi-legal option seems to be reinforced and sustained institutionally by local businesses, law enforcement officers, and local authorities because their interests are inextricably intertwined with it. Local businesses and law enforcement officers are the obvious beneficiaries of the rent paid by migrants; even the local authorities benefit from this form of rent as a result of the tax that is indirectly paid by migrants. By keeping the migrants' businesses semi-legal, the authorities have an excuse to crack down on migrants and their businesses whenever necessary. The whole situation constitutes a general equilibrium through which various interests are balanced.

This case study thus paints a far more complicated picture of the law's impact on people's behavior than is usually assumed. Instrumental concerns, or coercive action and sanctions alone, do not adequately explain people's interaction with the law in a “lawless” circumstance; a whole range of instrumental concerns must be considered, and they, together with sanctions, must be understood in the context of a larger institutional environment in which the interactions of various players unfold. Since there are both benefits and costs in the current situation for a wide variety of actors, the semi-legal behavior is reproduced in the society, even though law has become more and more important. It also becomes extremely difficult, if not impossible, to improve the situation. Obvious legislative prescriptions, such as removing the discriminatory legal barrier or improving migrants' rights, are not likely to occur without outside pressure or significant change in some economic factors.

I note that this case is by no means representative of the entire legal system of China. Not only does the interaction among migrants, locals, and state officials take place against the backdrop of the drastic developmental difference between rural and urban China, but the discriminatory treatment toward migrants itself is a result of China's current policy in preventing overpopulation in big cities. All these factors could distinguish this case from general legal enforcement in China. But I also note that the behavioral patter of the migrants is not so unique. As I illustrate, similar situations exist fairly frequently across the world.

The rest of this article begins with a brief description of rural-migrant entrepreneurs, most of whom are in family-based garment businesses, and how they have reacted to a set of discriminatory regulations in Beijing. It then examines the legitimacy of the law and explores the instrumental considerations of those who comply with and those who do not comply with the law, respectively. It finally focuses on those who take half-measures; that is, by renting licenses from the locals, and analyzes both the instrumental concerns of the migrants and institutional support from other players. This article concludes with some implications on legal enforcement in general.

The Case

Internal rural-urban migration has been significantly shaping the landscape of contemporary China's political economy. Statistics show that there were 80 to 100 million internal migrants in 1997, and 120 million in 2002;² this is larger than the total population of most countries. The numbers have been rising as markets have begun to erode the rigidity of the household registration system (hereafter termed hukou), which was strictly enforced until the late 1970s (Chan 1994; Cheng & Selden 1994; Dutton 1998).³ With the retreat of the Party-State and the economic reforms of the late 1970s (Oksenberg 2001; Peerenboom 2002:188–238), migrants have poured into cities in search of better living conditions and job opportunities. As massive migration floods the cities, however, the host governments in the cities have had a negative attitude toward the migrants, who are believed to be in competition with urban hukou holders for limited urban resources.⁴ For example, partly to prevent overpopulation and also to protect the interests of the locals, the Beijing municipal government has managed to control the flow of migrants through discriminatory legal treatment (He 2003a). Once again, the hukou system has become the institutional basis for migrant control. To stay in the city legally, migrants must apply for temporary resident permits (Ministry of Public Security 1985). To run a small family-based business, migrants must satisfy 10 requirements to gain individual business operator (getihu) licenses, six (see Table 1, Nos. 1–6) more than for Beijing hukou holders. These discriminatory requirements for license applications are clearly stipulated in more than 10 statutes and regulations issued by the Beijing People's Congress and the Beijing municipal government (Beijing Regulations 1995, 1–10).

Table 1. Documents, Minimum Fees, and Time Needed for a Migrant License Application⁵

Source: These fees and times were estimated from more than 50 migrants interviewed in Beijing in December 2002.

Note: 1. Yuan is the Chinese currency; at the time of this writing, one U.S. dollar was equal to 8.3 yuan; 2. The first six documents, which are more difficult to obtain, are extra or discriminatory requirements targeted at migrants only.

Despite these discriminatory regulations, more than 3 million migrants have resided in Beijing since the early 1990s (Beijing Statistic Bureau 2002). Although a large proportion of them have been performing dirty, difficult, and dangerous jobs that the urbanites do not want, some migrants have set up family-based businesses and accumulated a considerable amount of wealth, and have also occupied certain markets. For instance, migrant entrepreneurs, other than the locals, have dominated the garment market in the city. It is believed that migrants were the real force behind an overwhelming portion of garment retail stalls in Beijing. In a famous migrant community, Zhejiang Village, for example, more than 90% of migrants were engaged in family-based tailoring work, wholesale businesses, or retail sales (Wang 1995; Ma & Xiang 1998; Zhang 2001).⁶ These migrants rented houses from local peasants on the southern edge of Beijing, manufactured garments in sweatshops, and sold their products on the streets.⁷ It was estimated that by 1994, the migrants in this village alone controlled about 80% of Beijing's wholesale garment market for medium- and low-quality garments (Johnson 1994; Beja & Bonnin 1995).⁸ The commercial activities of these migrants extended far beyond this single locale in Beijing. Zhejiang Village has also been a major source of garments for northern China and even some regions in Eastern Europe and the former Soviet Union.

Most migrant entrepreneurs—the real owners of these petty businesses—conduct their business illegally, without possession of their own licenses as required by state and local legislation. According to an official investigation in central Zhejiang Village, for instance, 3,376 of the 3,724 businesses that produced garments did not have licenses (Xiang 1998: footnote 10). The findings of my own investigations were that from 1999 through 2002, almost all the stalls located in the basements of the garment transaction markets of the village did not have licenses. Whereas migrants were the real stall operators in well-known shops like Baihua and Wangfujing Store, in general they did not possess their own licenses. Investigations showed that many migrants actually conducted their own business although they rented licenses from local residents or used locals as a cover (Xiang 1996)—a strategy I call “legal collusion.” In legal collusion, the migrant entrepreneurs pay rent to locals in exchange for licenses and protection. The key to legal collusion is the collaborative relationship between ordinary migrant entrepreneurs and local Beijing business entities or local getihus, whereby locals act as the “front men” for migrants by being the official license holders. This collaboration is designed to evade license requirements. Both sides benefit from the collaboration: the migrant entrepreneurs enter the garment market with the protection of the locals, while the locals gain extra income from the migrants (He 2003b). At first glance, legal collusion appears to be corruption or legal evasion, but it differs from corruption in that few elements of state or officials are involved. In this case, for instance, migrants rely on locals—not on officials or other legal enforcement agencies—to avoid the license requirements. Legal collusion is also different from legal evasion, which usually refers to bluntly evading the law. The practice of legal collusion attempts to evade the law, but through subtle cooperation. Thus, some gray areas lie between legal evasion and corruption: if collaboration is with state agents, then it is likely to overlap with corruption; but if migrants completely ignore the license requirements and sell their products on the streets, it may be read as legal evasion.

Legitimacy

The previous section illustrates a situation in China where both locals and migrants flout, circumvent, or manipulate licensing legislation. A direct answer behind this situation could be that it is in their mutual interests to do so, and mutual interests arise only because of the introduction of discriminatory laws. If the laws in the city were not discriminatory, then there would be no incentive to engage in such a collaborative relationship. It is the existence of the “shadow of the law” that gives rise to these practices (Mnookin & Kornhauser 1979). As many studies clearly indicate, due to the existence of legitimacy of the law, many people will obey the law even though this contradicts their interests (Tyler 1990). How do migrants view the legislation in this case, or, to put it another way, to what extent does the law hold legitimacy?

Legitimacy here means the respect that people show to the authorities, or the sense of obligation to obey the authorities, or the reservoir of loyalty that the authorities can draw on (Tyler 1990). In itself, the law is not automatically legitimate. In modern states, as claimed by Weber, most laws are legitimate because they are expressed in a “system of consciously-made rational rules … which meet with obedience as generally binding norms” (1954:336). Other theorists further suggest that the law needs to “create the shared sense that existing legal arrangements are as they ought to be,” a process Sarat calls “legitimation” (1993:662). Yet to what extent do the empirical materials here fit into these theories, or do they suggest another story? As I shall demonstrate, the way in which the law originated and is enforced leads to a lack of legitimacy amongst migrants.

How the Law Originated

The Beijing municipal government's efforts to employ the law in governing and controlling migrants are mostly responses to the interests of urban bureaucrats and urban registered citizens. Although governments in China do not operate as do Western democracies, nonetheless, they face social pressure from their constituencies. Given the current political system, and especially the hukou system, urban governments more or less represent the interests of local hukou holders. Moreover, while the Beijing municipal government has consulted representatives of the local congress and local organizations (Xiong 1995), there has been no indication that the government has ever consulted the migrants themselves.⁹ To millions of migrants, at least, the legislative processes for these laws and regulations are hardly rational. The migrants live and work in the city but they have little say in the laws and regulations; this clearly aggravates their already difficult lives. The laws are, in large measure, cooked in a bureaucratic kitchen where the migrants basically have few means of making themselves heard.

The way in which these laws have been enacted has direct implications on their legitimacy. It is unlikely that laws made in this way will gain support and legitimacy amongst those people affected (Luhmann 1975). As I show below, most migrants are indifferent to these laws.

How the Law Is Enforced

Most times, the discriminatory laws are not strictly and consistently enforced. Instead, sporadic campaigns are launched to enforce the laws during politically sensitive periods. Other than these times, there are no campaigns and the laws become more or less “dormant.” This subsection first shows what happens to migrants during these politically sensitive periods.

Campaigns are usually launched in the name of “cleaning up and reorganizing” (qingli zhengdun): to combat unregistered migrants, unlicensed businesses, and illegal housing compounds.¹⁰ The scale of each campaign varies depending on the state's commitment to pursuing the particular campaign. Campaigns take place more often during politically sensitive periods or festivals, such as the anniversary of the 1989 Tiananmen crackdown on June 4, the national anniversary of the PRC on October 1, and the meeting period of the national people's congress and the national people's political consultancy in March. Some campaigns are organized by district governments, while others are launched by the municipal government.

The situations faced by migrants can be illustrated by the following narratives.

A migrant recalled:

I was walking on the street, on my way to buy some daily items. All of a sudden, some guys in uniform appeared in front of me and asked me for temporary resident permit. I did not have one at that moment. They just stuffed me into their car. I still remember that those guys pushed me so hard. Did I not know how to walk?! They sucked. Then I was taken to the police office at Dahongmen. Some 15 migrants and I were locked up in a small room. No place to sit and the smell there was terrible. During the first night, the police did not ask us questions. Some guards there, hired by the police … brought some chairs to female migrants. The next morning we begged the guards to buy us some food as breakfast. When we needed to use [the] toilet, someone just monitored us. Some of us used the phones of the guards to call friends. Some of them were released after their friends outside found connections and paid the fine.

I was not that lucky since my friends did not have effective connections. I then was taken to Changping,¹¹ where a bigger labor reform center was located. A policeman asked me a lot of questions and I was forced to sign the answer sheet. Later on they decided to send me back to my hometown. Of course, I was forced to pay the train ticket. It was such a scary experience. (Interview with a female migrant on December 15, 2002)

Another migrant entrepreneur recounted his experience as follows:

The local public security carries out inspection periodically around here to check whether everyone has got the temporary resident permit. If you do not have the permit, you better be smart and beg for mercy. Then they may let you off with a light penalty. If you dare to talk back, boy, you'll be dragged back to the police station to be beaten up.

You see, the public security has a quota of arrests to meet every month. If they have difficulty in meeting the quota, they would just randomly pick up some persons on the street to meet the target. If the public security catches a bad guy, he can walk away with the proper bribe. The public security has middlemen in the street to accept the bribes. (Wong 1998:104)

I once asked Deputy Luo, a policeman in his early 30s, this question: Do the police really have a quota for arrests?

Luo: Sometimes. For example, in 1999 when Beijing was going to have a grand ceremony for the nation's 50th anniversary, every police office at district level was assigned a quota. I was then working at a suburban prison, but I was summoned to downtown areas to help other fellows. Police then checked migrants' temporary resident permits. And we sent them back to their hometown. Our office was rewarded 100 yuan for sending each migrant back to his or her original place. Admittedly, not everything was conducted 100% legally. I also heard that some policemen tore migrants' temporary resident permits apart. (Interview on July 27, 2002)

In which case, what happened during normal periods when there were no campaigns under way? A migrant complained about his experience:

They (the government and its officials) have different categories of temporary resident permits, A, B, and C.¹² If you are holding a B or C permit, they know you are not powerful and dare to tear up your permit. Last time I was carrying my goods with a tricycle cart, and they were checking on the crossroads. They asked me for the temporary resident permit first, and I got one. And then they asked for the green card, and I also got one. They then asked for license for the tricycle cart. Come on! You needed to have a license for those tiny carts? I pulled it with my feet! They said, well, if you did not, 20 yuan for fine. If I do have a license for the cart, they would say that I overloaded. This was happening to my pal. He got all the permits and licenses but still got fined for overloading. I just sold two pieces of garments in one morning, but those who are in charge of tricycle carts have a far more wonderful “business.” Doing business here, we are sending money to Beijing. But they do not protect us. On the contrary, they fine us.

If you got money, everything will be fine. If you do not, no matter how many licenses and permits you have, they are still going to round you up. Take the temporary resident permit as an example, it is said that five yuan will do. But they just take 40, 50, 80, or 100. When we use a tricycle cart to carry garments, we get fined all the time. The guards take the fine, and the police are standing by. Twenty yuan! And no receipt! And if you want one, they said the transportation team [jiaotong dui] would issue one. But who would travel to the team [to] get a receipt? Doesn't it involve energy and time? And I do not know where the transportation team is!

The other day my father was playing chess at the No. 17 bus station with his buddy. He did not have his temporary resident permit with him that moment and was rounded up immediately. He was soon brought to Changping. We happened to have a friend in the government and the friend paid a visit and brought a packet of cigarettes. It was said that everything was fine. But my father was taken to Xuzhou¹³ the next day! My brother and I spent a total of 300 yuan to get my father released. Not to mention other losses! (Interview with a migrant on December 14, 2002)

Another migrant entrepreneur told me:

There are numerous examples of breaking the law and I do not know where to start. While Zhejiangren (people from Zhejiang) are not necessarily law-abiding people, they are not addicted to breaking the law. Under many circumstances, they do not have a choice. For example, when ICB [Industrial and Commercial Bureaus] guys dropped by, they initially asked for 2,000 yuan for fines. After bargaining with them, the fines were reduced to 1,000, or simply 500. There is no uniform standard in this regard: at one place, a sewing machine is fined 100 yuan; it is 500 yuan at another place. If you refuse to pay, ICB guys will take your garments or sewing machines away. But if you send their boss some money, a carton of Marlboro cigarettes, or a leather jacket, everything is OK. Last year my business was not good, and I refused to pay fines. An ICB guy insisted that I should at least pay fines for one of my three sewing machines. I eventually gave him 100 yuan. He immediately left without even issuing me a receipt. The money must have been swallowed by the bastard. (Interview on July 29, 2002)

When asked why the migrants did not try to understand the regulations and apply for a legal license, most migrant entrepreneurs were surprised by the question and would then answer, “The result is the same, no matter whether you have the license or not; why bother to apply?!”

Out of curiosity, I asked Boss Shen why “the result is the same.” He hired three workers to manufacture garments in his two-story rented apartment. A signboard hanging on the door read “Golden Sheep Leather Garment.” As required by the relevant laws, the owner should hang the business license on the wall, but I did not see a license inside his apartment.

Boss Shen: Well, the current practice is called “replace fees with fines” (yifa daijiao). If you [referring to migrant entrepreneurs] do not have a license, the taxation bureau and the ICB would simply come periodically to take fines. After you pay the fines, they would be happy and leave.

Author: What if they come all the time?

Boss Shen: Up to now, they did not do that. If they do, then I have to shut down my business and go. Currently, the fines are more or less the same as the fees that we shall pay according to the laws. (Interview on July 29, 2002)

One might think that the officials' greed can never be satisfied, since constraints such as checks and balances are not available. Boss Shen's testimony, however, reveals that this is not the case; while officials may cite several legal regulations in order to impose a fine, the total value of fines does not go beyond a certain limit and is actually more or less equal to what the law demands.¹⁴ There is a similar situation in the issuance of temporary resident permits. Since 2002, the fee for the permit has dropped from 360–400 yuan per year to five yuan per year.¹⁵ It was assumed that most migrants would apply for the permit because of the drop in the fee, but the situation has not changed much. A migrant involved in the clothing retail business in Zhejiang Village said:

I have been in Beijing for more than a decade. Originally, to apply for a temporary resident permit demanded more than 100 yuan. Later on, the newspaper said it has been changed to five yuan. But you know that there is policy above, there is counter-strategy below. Some local governments simply do not follow the new policy. In Fengtai District, on the first day, you may get the permit for five yuan. On the second day it becomes 60 yuan. And third day, 80 yuan, and fourth day, 120 yuan. No receipt; no receipt even for a five-yuan permit. Where can you file your lawsuit? I have asked a lawyer regarding this. The lawyer said a lawsuit may begin with a receipt, otherwise the courts won't take your complaint. Right, it is the administrative litigation you just mentioned. The court will listen to the government; they are combined together. And all these are orchestrated by the Fengtai government. Who can win the case? If we file a lawsuit, they [the government] will round you up for 15 days, and claim that you are violating social order. Now they require us to have a green card. The temporary resident permit is not enough without the green card [the migrant displayed the green card]. This card will take 200 yuan as deposit. But I never heard of the deposit being returned. No green card, no resident permit. It is said that in Siliuyuan¹⁶ the permit only costs five yuan. But I spent 60 yuan for it. They do check whether we have the permit, coming here with their 110 police car.¹⁷

Not everyone can get a temporary resident permit even if you voluntarily apply for one. The police may simply refuse to issue to you, without giving any reasons. But when you are caught without the permit, they would impose a fine on you. According to what I have heard of, the policemen would check your temporary resident permit when they bump into you. If you did have one, they might tear it apart, and then ask you for the permit again. What can you say? What can you do? In a word, everything is up to them. (Interview on July 26, 2002)

In reality, if caught by a local policeman for not having a permit, migrants will bribe the policeman in order to remain in the area. Some might be caught several times a year. The fine usually does not go beyond 50 yuan at a time so that overall, it is still more or less the same as the fees migrants would have to pay if they went through all the legal procedures (Zhang 2001). In other words, the fines do not completely destroy migrant businesses. Instead, the officials intentionally allow some space for these businesses to grow. The reason is that if the businesses do not survive and the migrants eventually go out of business, the officials will lose opportunities to extract benefits for themselves. Thus, the fines do not exceed a certain amount and can be adjusted through bargaining, bribery, and negotiation. Officials thus artificially create opportunities to reap benefits by manipulating the regulations, a phenomenon called “rent-seeking” in the literature (compare with Krueger 1974; Bhagwati 1980).

The situation mentioned above indicates that the law is manipulated by the state and its officials so as to achieve multiple goals: to clear and control migrants; to line their own pockets; to create a constant sense of anxiety about the exercise of state power; to perpetuate uncertainty and, therefore, provoke migrants to feel vulnerable to law enforcement; and to allow space for migrant businesses to grow. In this instance, the law suggests a confluence of formal and informal rules, and the migrants can develop their businesses with “certain” certainties; yet their development is no more than that of a bird in a cage (Lubman 1999). The laws are sporadically enforced and sometimes manipulated by the state. This seems to confirm what Packer maintains: the lack of legitimacy is further aggravated by the sporadic way in which these laws are enforced (1968:287). As a result, the legitimacy of the law is undermined rather than enhanced.

Migrants' Perception of and Reaction to the Law

A cycle rickshaw–puller in his 40s told me of his impressions toward law enforcement staff:

The legal enforcement team (zhifadei) acts just like the Japanese invaders (riben guizi). While they do not commit arson, murder, or rape, they do have a “three clear” (sanguan) policy to us.¹⁸ They take away whatever I have—that is, my money, my personal identity documents, and sometimes my cart. I could do nothing to them; they have power while I do not. (Interview on July 26, 2002)

Another male migrant entrepreneur said:

Those government announcements are just like farting. They said that the illegal marketplaces would be demolished in June. But now it is already December. Guyuan, Tianhai, which marketplaces have been demolished? Money is the only thing that matters. You know what, keeping these marketplaces equals money. They just want money; they want nothing other than money. Nowadays no one dares to speak out truth; I dare. What are you really doing here? Did the Fengtai government send you here? Let me see your identity card. If I tell this to a reporter and my conversation appears in the newspaper, they will round me up.

Beijingers are so bad. The Fengtai district government is corrupt to the root. I rent a place from Bejingers. Some Beijingers are bad and some are OK. Anyway, the police are the worst. In the marketplace over there, more than 10 stalls next to the streets are owned by policemen. But we still need to have a good relationship with the government. Otherwise when something happens, it would be too late. (Interview on December 8, 2002)

Boss Wang, after learning that I was interested in how the migrants deal with officials, became enthusiastic to tell me his story:

There is no secret inside the game; you can wrap up your investigation after I tell you this. If you use others' licenses and get caught, the first thing you need to do is to find guanxi [connections]. Through the connections, you must send the officials gifts. That is why we have to cooperate with Beijing locals: they are the first connection for us. After that, you will get to know the officials, and you better keep the relationship. You may want to send more gifts during festivals to stabilize the relationship. Do you not understand today's logic? Money can do everything. Otherwise, your troubles never end. (Interview on July 26, 2002)

Migrants' distrust of urban authorities can be easily seen from their responses to the following question. When asked if they would report their cases to the police if they were robbed, 90% of migrant respondents said yes, but 80% of them immediately added the clause, “I do not think the police will take care of it, though.”¹⁹

Consequently, migrants rarely resort to the law to protect themselves. At the start of my fieldwork, I was surprised to find that in most cases, the migrants did not even know the regulations that substantially affect their lives. I once showed them a manual containing the series of regulations and laws promulgated in 1995, but none of them recognized the laws. One of the reasons for this could be that the migrants do not have sufficient resources or energy to understand each and every regulation. The various legal requirements have become a labyrinth that they cannot figure out. There are, however, more important reasons for this “ignorance of the law.”

A migrant entrepreneur said:

These laws are useless to us, little brother. You are a law student, so you may know how to use the law. But we do not. We do not even know where to sue them [the bureaucrats who infringe on migrants' rights]. Moreover, those officials are all in cahoots together. How could they speak out for us? (Interview on July 27, 2002)

China has been implementing its Administrative Litigation Law (Xing zheng susong fa) since 1990. The migrants inside Zhejiang Village, however, rarely initiate litigation against the state. When I asked why they do not file their complaints in court, some migrants answered, “We do not know how to sue them, and we do not want to be troubled. We would rather pay fines and move on.” In response to my question “If you have all the required documents, including the business license, would that make you safe?” they replied:

That will not make a difference. If you have document A, they would ask you for document B. If you have license B, then they would ask you for document C. If you have all the documents, they would say that your documents are not updated. They can always find some defects that do not 100% conform to the law. I am not an expert in law, so how can I know whether what they say is true or not? In any event, what they want is money. If they get the money, everything is OK. If not, everything is useless.

It is pointless to argue against them with the law. Even if you win the case this time, you are still subject to the control of these officials. They will give you a lot of trouble later on. As long as you still want to do business here, never think about striking out against a rock with an egg, never. (Interview on July 26, 2002)

These views from migrants were further substantiated by an official:

Luo: There are more and more administrative lawsuits against us. A few of them are from migrants. But generally speaking, migrants rarely sue. If they do, there must be a white-collar spouse or relative behind them. In other words, only those people dare to initiate the proceedings. On the state side, usually courts are cooperative with us, unless the police have obviously done something wrong. Usually we do not lose litigations. When it is likely that we are about to lose, we will use alternatives, such as mediations, to negotiate with litigants. (Interview on July 27, 2002)²⁰

Since using the law and confronting the state or officials are not effective or even feasible ways to protect themselves, migrants have to take up the weapons of the weak (Scott 1989). It is not surprising that when migrants get caught for not having a license, they do not defy the authority of the state. On the contrary, they often act with as much humility as possible so that officials may reduce the penalty out of sympathy. Once the migrants are released from a round-up, they will return to their businesses in the original, collusive manner. The same situation is exemplified during the sporadic campaigns. From their experiences, migrants come to realize a campaign will end sooner or later and that everything will return to normal after the campaign is over. Thus, they simply leave Beijing when the campaign is at its height and return when it is over (Solinger 1999:69). To avoid potential losses in the future, the migrants find better and more creative ways to hide their businesses, and to make the campaigns less effective (He 2003c).

I asked Deputy Luo, “Some migrants told me that they returned to Beijing earlier than the police officer who deported them came back. Is this true?” His response was as follows:

That happened. During sensitive periods like 1999, what we could do was to send some migrants back, so as to restore the order for a while. After the sensitive periods, we do not have energy to round up these migrants. At the end of the day, migrants are only making a living here. Moreover, we also have to cooperate with migrants. There are some wealthy getihus here, behaving arrogantly because they have money (caida qichu). But our police always lack enough financial sources. We, thus, have to rely on them for donations, so we all have to be tolerant. Anyway, we have to regulate some activities, for example, clan fights, rapes, homicides, or drug trafficking. (Interview on July 27, 2002)

Three points can be inferred from Deputy Luo's testimony. First, migrants are unable to use the Administrative Litigation Law to constrain officials' behavior or to protect their individual rights. Administrative institutions, including the police, are closely entwined with the courts. Second, the enforcement of law is quite arbitrary, including acts such as tearing up migrants' temporary resident permits and randomly rounding up migrants; but this largely happens during law enforcement campaigns, or during sensitive political periods. Third, the law is enforced differently for different types of migrants. For poor migrants, the law is used to extract fines and deport them so that the police can claim rewards.²¹ As for wealthy migrants, however, the local police are more interested in building up a patron-client relationship with them to obtain donations for themselves. The police have a collaborative relationship with wealthy migrants while they also rely on them.

Migrants altogether feel quite useless in understanding and using these laws. The officials have arbitrary power to interpret and enforce the law, even though they may choose to allow migrant businesses some space to develop. Ultimately, migrants are still at the mercy of officials; if the officials cannot punish the migrants today, they can always do so tomorrow. For many migrants, the law is more of an unofficial way for the police to create an income on the side rather than an effort to keep public order and restore justice for migrants. The regulations of the state are, therefore, simply a sham, a pretext for government officials to prey on migrants without delivering any legal protection. The law has become a scheme through which these officials dupe migrants to extract benefits for themselves; migrants are free to do anything but use the law to protect themselves.

In sum, the law lacks legitimacy for a variety of reasons. It appears that those laws made without general consultation with affected groups may lack legitimacy. Sporadic and arbitrary enforcement further undermines legitimacy, which resonates with the proposition that procedural justice is a key antecedent for legitimacy (Sunshine & Tyler 2003). The migrants generally feel powerless against the state and officials, and they do not have an incentive to know the law.

The Costs (Instrumental Concerns) of Legality

Legitimacy is widely held to be the key to the success of legal authorities. “If authorities have legitimacy, they can function effectively; if they lack it, it is difficult and perhaps impossible for them to regulate public behavior” (Tyler 1990:57). To what extent is this assertion true in the case of China? How exactly do people decide whether or not to comply with “illegitimate” laws? Starting with this section, I compare three strategies that migrants use in dealing with “lawless” or “illegitimate” discriminatory regulations. Broadly speaking, migrant entrepreneurs can be divided into three categories: those who comply, or at least do so to the best of their ability; those who do not comply at all; and those who engage in legal collusion by taking half-measures. I expected most migrants to make a choice primarily on the basis of instrumental concerns, as suggested by studies in post-Communist countries (Feige 1997; Leitzel 1997), but this hypothesis was only confirmed in the first two categories—legality and illegality; in legal collusion, things are far more complicated.

My approach is enlightened by, but different from, that of de Soto (1989). In his study, de Soto divides the costs of formality into “the costs of access” and “the costs of remaining” in it to examine all the requirements one must meet in order to legally enter a specific economic activity, and the requirements one must meet in order to preserve legality, respectively (de Soto 1989). I estimate the costs and other instrumental concerns of a migrant business in Beijing fundamentally through interviews.

The Costs of Access

According to local legislation, six documents have to be presented to the Beijing ICB as part of a getihu license application. As illustrated below, getting these documents and going through the application process involve a lot of energy, time, and costs for migrants, especially when the officials in charge enjoy a wide range of discretion.

The document “Lawful Evidence of a Fixed Business Site” is viewed as the greatest obstacle for migrant entrepreneurs. Most migrant households work from a room rented from local peasants, but rental housing from local peasants is usually not considered by the Beijing ICB to be “fixed.” Since the small scale of migrants' petty businesses does not allow them to rent long-term commercial housing in the metropolitan area, where prices are prohibitive, it is almost impossible for migrants to obtain lawful evidence of a fixed business site. This requirement constitutes an almost insurmountable barrier for those who want to have a formal stall on the streets, and it is for this reason that collaboration with local businesses was initially adopted (Ma & Xiang 1998).

A temporary resident permit is a prerequisite for license application, but the application for the permit involves much hassle. For the application, migrants are required to present two certificates: a permission slip for leaving their hometown, issued by migrants' original township governments, or the letter of introduction issued by the ICB in their hometown; and evidence of registered housing in Beijing. Generally, the most important criterion for granting a temporary resident permit is whether the migrant lives in housing registered under Beijing landlords. According to the law, a landlord is supposed to take his or her migrant tenants to the township government to get a green card—a certificate that indicates acceptance by the government. Although there are no regulations to stipulate that governments can charge for the issuance of green cards, they still charge 50 yuan for each card. Only with the green card can migrants apply for temporary resident permits at the police departments.²² In reality, however, police departments only accept landlords' rental agreements, which prove that their rental housing is registered with the Housing and Land Bureau. To register rental housing, Beijing landlords have to apply for a rental permit (zhunzuzheng), which involves an application fee that reduces their profit from rent.²³ As a result, Beijing landlords are usually reluctant to apply for the permit, so the majority of migrant entrepreneurs cannot present the police with registered rental agreements. It is not surprising, therefore, that applications from those migrants who have not yet rented registered housing are usually denied.

The practice of granting temporary resident permits has nonetheless varied over time and among locations in Beijing.²⁴ Although the law demands that all migrants must register, not all migrants can obtain temporary resident permits even if they voluntarily apply for one.²⁵ Local criteria decide who is deemed a desirable or undesirable migrant, and these further determine whether a migrant has the right to register for and stay in a given community (Zhang 2001:35).

Female migrant applicants must present “the certificate of marriage and fertility card” (hunyuzhen) to show that they have taken birth control measures. As migrants are perceived to have a high fertility rate, officials are overly concerned with “excessive birth” (chaoshen) among the migrant population and take intensified measures to monitor migrant fertility practices (Davin 1998).²⁶ In practice, the Beijing Birth Control Committee requires female migrants to update their fertility history twice a year with their hometown government, which is impractical for most female migrants.²⁷ As a result, many female migrants simply ignore the requirement and pay a fine if caught (Zhang 2001:35). Nonetheless, those who do have the card claim that they paid about 20 yuan on average to get the certificate, and 50 yuan for the physical examination.²⁸

Moreover, it should also be mentioned that the resident permit has to be renewed annually. Since a temporary resident permit is only valid within a particular district, a migrant who moves to another community has to apply for a new permit. All these factors account for why so many migrants still do not obtain a permit even though the “official fee” for it has been just five yuan since 2002!²⁹

Only with all the first six documents listed in Table 1 can migrants eventually apply for a getihu license from the Beijing ICB. Since the process is so complicated, few migrants have a clear idea of what the procedures and requirements really are. This gives the ICB a good deal of discretion on the matter of issuing licenses. There is an apparent information asymmetry between migrant applicants and ICB officials. Partly due to this institutional design, and also due to the general expectation that officials are corrupt, bribery is a common feature in the whole process (Manion 1996). To make the situation worse, whether or not licenses are issued to migrant entrepreneurs depends heavily on the general political climate concerning private enterprises and migration. Migrant entrepreneurs face not only the general restriction on private enterprises, but also discrimination regarding their rural status. Thus, they suffer a double whammy in the application process (He 2003b).

With regard to how much effort migrant entrepreneurs have to put into getting a license, and the length of time they have to wait, almost every migrant has a different story. A migrant mentioned that it took him more than six months to get the license, after he visited a local ICB about 30 times, spent more than 3,000 yuan, and got to know every official at the ICB.³⁰ Other migrants, however, said that they only spent 10 days and several hundred yuan to get a license. The duration for each approval or registration in 2002 is indicated in Table 1. On average, it is estimated that the cost of applying for a getihu license lies somewhere between 1,000 and 5,000 yuan, and it generally takes 60 to 100 days to get through the bureaucracy.

The Costs of Remaining Legal

Once migrants enter into business, they have to comply with a number of regulations in order to remain legal. The “costs of remaining legal” include tax costs and ICB regulation fees. For the garment business, the taxes and fees have different rates at the manufacturing stage and selling stage. At the manufacturing stage, the tax is 100 yuan per sewing machine per month, and the business regulation fee is 30 yuan per person per month. At the selling stage, the taxes and regulation fees are set at 5% and 3% of business income, respectively (Beijing Regulation 12 1995). However, as most migrant entrepreneurs do not keep accounting records and, in any event, officials do not trust migrants' records, officials usually estimate and impose a tax and fee based on the size and street locations of the garment retail stalls. Sometimes, a stall of comparable size and location is used as a reference.³¹ As there is no clearly defined standard, officials enjoy a great deal of discretion even though, as indicated earlier, their greed might be self-constrained. Migrants often offer them bribes to influence their decisions.

To remain legal in the garment business, migrant entrepreneurs must apply for work permits for their employees. To obtain a permit, both the employer and the employee must register at the street neighborhood committee and at the Labor Bureau. The employer has to apply for an employment permit (gugongzheng), while the employee must apply for a work permit (wugongzheng).

According to the Beijing experimental measures on management of the laborer employed by getihu and private owned enterprises (Beijing Regulation 11, issued 1989, amended 1997), employers cannot demand that their employees work more than eight hours per day without employee consent and without giving overtime pay. This contradicts the common practice in Zhejiang Village, where migrant laborers usually work 16 hours a day during the busy season (Solinger 1999). Moreover, the Beijing standard of food and housing for migrant laborers, issued on June 27, 1995 (Beijing Regulation 12), requires employers to provide two-and-a-half square meters of housing per employee, which is impossible for migrant entrepreneurs in Zhejiang Village to comply with.

It is, therefore, prohibitively costly and inconvenient for an internal migrant to conduct business in Beijing in full compliance with the law. In comparison, taxes and ICB regulation fees constitute only a small portion of the whole cost of legality.³² These complicated processes seem to make a huge difference in the migrants' choice between legality and illegality. Only those who are able to deal with the legal requirements and handle the costs will consider the legal choice, and eventually, only those who believe that the legal option will offset the costs of illegality will choose the former. In reality, only a slight portion of migrants take the legal option, and they choose legality largely because their businesses are generally large in size. When asked why they have satisfied the requirements that most of their peers did not, they said that they wanted as little trouble as possible. Unlike what Tyler (1990) has found in the United States—that legitimacy plays a significant role because people will obey the law even though they think that the law or the enforcement of the law is not right—I did not find that the group that chose legality, as compared to other groups, perceives the law as being more legitimate. Two reasons seem plausible for the decision: their businesses have grown to such a size that it is impossible for them to hide from local authorities, and the size of and profit from their businesses allow them to pass though these endless bureaucratic hurdles.

The Costs of Illegality

After examining the costs of legality, this section moves on to illegality, which primarily takes the form of “street vending and escape.” When migrant entrepreneurs conduct businesses in this manner, they do not possess a license, not even one that is borrowed or rented from others. Neither do they go through all the troublesome procedures mentioned earlier nor pay the cost for being legal. There is no doubt that this group benefits economically, which may partially offset the cost of illegality and even outweigh the benefits of legality. However, these migrants are not necessarily better off; being illegal always involves some inevitable costs. There are direct costs of illegality, and indirect costs as well—the benefits they have to forgo when they choose illegality.

The Direct Costs of Illegality

The most significant direct costs of illegality are the enormous investments migrants have to make to avoid detection. They constantly run the risk of being penalized for not having permits, paying taxes, or applying for the authorization required by law. They save on the cost of compliance but must bear the cost of avoiding corresponding penalties or even confiscation of their garments. Fieldwork data show that migrants have to be on the alert for officials when they engage in “street vending and escape.” They have to be alert when selling their garments on the street and adopt strategies to avoid detection when producing garments in rented housing. This not only involves certain human resources but also affects their choice of business locations (Smart 1989:60). When these migrants choose to rent a location, safety is usually an important consideration, so the places they rent may not be the best in terms of economic efficiency.

A senior male migrant entrepreneur said:

By then [in the early 1980s] Beijing took very strict control over business activities conducted by outsiders, especially in the downtown area. There were various street patrols every day. Whenever the street patrols found people from outside Beijing conducting business, they would detain the outsiders for a while, and then deport them out of Beijing with their goods confiscated. My brother and I were renting a room in the downtown area. We were detected by city patrol, detained, and deported a month after we arrived. But we did not go very far and stayed in Shijianzhuang [the capital city of Hebei province, about 200 miles south of Beijing] region for half a month. Nonetheless, I thought that Beijing had a big market, and it was easier to conduct business there. So we came back to Beijing again. As we had already learnt the lesson this time, we dared not rent a place in the downtown area. Instead, we walked in the direction of the outskirts. We later moved frequently to avoid being detected. (Wang 1995:35)

The venue for garment retail is also affected, as migrants have to choose locations that officials rarely visit; these locations are usually not the best thoroughfares for conducting business. They also have to reduce the duration of street vending to lower the risk of being caught. Some choose to operate after business hours, especially during summer when it is light until seven or eight in the evening.

Further, the scale of manufacture has to be kept within a certain level so as to avoid discovery. Sometimes, migrants deliberately limit the scale of their operation. If they have to grow, they do so by dispersing their employees to many establishments, thereby increasing the already prohibitive rent. While such arrangements are obviously helpful for avoiding detection, they also prevent these migrants from achieving efficient scales of production.

Moreover, when migrants conduct “street vending and escape,” they have very limited ways of attracting customers. For example, they are unable to rely on advertisements, which could help offset the disadvantages of a poor location and compensate for the lack of visibility.

Since these migrants cannot access formal marketplaces, the price of the garments they sell on the streets is much lower than those in formal shops or stalls in markets, even though there is no difference in quality. Customers believe that the quality of goods on the street is poorer than those in formal markets and are aware that migrant sellers will accept a lower offer.

The Indirect Costs of Illegality

By choosing illegality, migrants have to bear not only the direct cost of illegality but also the cost of being unable to employ arrangements such as sales contracts that depend on the legal system to be implemented (de Soto 1989). A contract is an institution that binds different parties to an agreement. A legally enforceable contract creates certain expectations, encourages the parties involved to make reasonable commitments they can fulfill, and discourages unrealistic pledges by specifying the penalty that the defaulting party must pay. Contracts facilitate transactions in social-economic lives. Migrants who choose illegality largely cannot access contracts because they are absolutely unwilling to expose the illegal nature of their businesses in court, an important component of local authority. Subsequently, many negative consequences may result. The migrants' unwillingness to go legal will make agreements among them, whether written or oral, basically unenforceable. Due to the lack of enforceable contracts, some transactions may never take place. For example, when an illegal migrant entrepreneur wishes to buy raw materials from a supplier, the supplier cannot be sure that the migrant entrepreneur will honor the price they originally agreed on. The supplier might then agree on transactions with the migrant only if the profit for the supplier is sufficiently large. These considerations will probably thwart many otherwise feasible transactions.

Moreover, the lack of enforceable contracts makes it difficult to procure beneficial long-term commitments (de Soto 1989). For instance, if a migrant entrepreneur is assured by contract that a customer will buy a given quantity of garments over a period of time, he or she will then be able to order raw materials, hire migrant laborers, and invest in the necessary sewing machines. This cannot happen for migrant entrepreneurs operating illegally because they lack the power to bind by contract.

In addition, the scale of production is limited due to a lack of contracts. Under completely illegal circumstances, migrant entrepreneurs never know if their buyers will show up as promised. Thus, it is highly risky for migrant entrepreneurs to increase the amount of products they sell every day. Migrants who conduct street vending without licenses do not enjoy economics of scale.

While many informal systems have been creatively generated by migrants to cope with the inability to access the formal legal system, they do not function as well as an efficient legal system (de Soto 1989:165–6). One way to reduce the possibility of breach of contract, for example, is to only do business with people from the same towns or villages (He 2003a). If any party fails to fulfill his or her contractual obligation, the innocent party will go to the family, relatives, or friends of the defaulting party in the hope that they will exert pressure on the defaulting party to honor the contract. The problem with this alternative, though, is that it limits transactions to a narrower circle, so the scope of business will be similarly limited (de Soto 1989; Landa 1994).

Another alternative used by migrants is to invest time and energy in cultivating long-term business partners. For instance, a garment manufacturer may want to establish relatively stable relationships with those migrants who rent counters or stalls in marketplaces (Xiang 2000). This is beneficial to both sides, because the renter wants to have a constant supply of garments to sell and the manufacturer needs a channel through which to sell the garments. The penalty for noncompliance is to disclose the violation of contractual obligations to mutual friends or relatives. This serves two goals: it damages the defaulting party's reputation, and it isolates the defaulting party in the community so that others will not conduct business with him or her in the future. This is quite effective in ensuring compliance with contracts because personal reputation is an important asset for migrants who live in a relatively small enclave where people have to rely on one another.

There are, however, certain limitations to this alternative. First, it takes a lot of time and energy to form sustained relationships. Second, those who are new to the market will find it hard to be accepted as business partners. They will begin to establish mutual trust only after completing an appreciable number of transactions. Until then, newcomers have to pay a high price to secure contracts since others will be afraid of becoming the victims of a breach of contract. In fact, fieldwork data show that migrants will give priority to established customers.

Since a good reputation cannot guarantee that a party will not breach a contract, migrants tend to spread the risk by dealing with many customers. For example, even when a migrant entrepreneur obtains a big order from someone outside the migrant enclave, he or she usually collects garments from individual manufacturers first and then uses all these garments to fulfill the big order. If the entrepreneur hires more employees to satisfy the order, the business will suffer if the potential customer does not show up as promised. Apart from avoiding detection, this is another reason why manufacturers usually keep their production scales low, resulting in higher unit costs and reduced profits. To ensure that their contracts are honored, migrants may resort to threats or, occasionally, violence. This is, of course, not an efficient way to conduct business. To avoid becoming the target of violence, some migrants pay protection fees to underground organizations (He 2003c; Xiang 2000).

In sum, whereas illegality saves on the costs of legality, illegal migrants are constrained by a wide range of shortcomings, including a relatively thin profit, unfavorable operation locations, fewer hours of operation, poor labor utilization and small scale of operation, and no access to the formal legal system. Their businesses are regularly interrupted by raids. If they are caught, they suffer significant economic losses on their goods, not to mention the hours of labor and other costs. Not surprisingly, this manner of conducting business was only popular among migrants from the early to mid-1980s, when most migrants had just arrived in Beijing and could neither pass the bureaucratic hurdle to become legal nor establish collaborative relations in legal collusion. Currently, some migrants still conduct business in this way, and they are mostly newcomers to the city, or the mobility of their small businesses allows them to do so economically.

Why Legal Collusion?

This section focuses on the following question: Why is legal collusion widely adopted by migrants, yet mostly tolerated by the authorities? In this section, I suggest that the instrumental concerns of migrants are important but not adequate in answering this question; the advantages in social norms that migrants gain in legal collusion and, most important, institutional support from various actors are crucial in widespread legal collusion.

For migrants, legal collusion definitely involves certain costs. For example, in counter-renting—a popular form of legal collusion—the most obvious cost is the high rent that the migrants have to pay to the locals. The rent includes the cost of legality that the locals have to pay as well as the profit the locals can gain from this collaboration. When locals apply for a license, they must pay regulation fees and taxes. As a result, the rent that migrants pay to locals has to be much higher than the original fees and tax. The extent of profit that the locals can make largely depends on the market's demand for and supply of counter spaces. Furthermore, while counter-renting offers camouflage that is difficult for the ICB to see through, it still involves some costs for avoiding detection. The usual practice is for the local cooperator to show up at the counter for an hour or two every working day; this too is a cost. When issues arise in the rented stalls, the locals who hold the license have to deal with the officials. Certainly, the costs are eventually borne by the migrant and, if the collusion is detected, the migrant entrepreneur may have to bear the penalty, including a revocation of the local's license and confiscation of the migrant's garments. To avoid big losses, the migrants' scale of production has to be kept small.

In comparison to legality, counter-renting does not require all the procedures that are targeted at migrants, since the locals are responsible for applying for a legal license and dealing with unexpected harassment from officials. This saves migrant entrepreneurs a lot of money, time, and energy. Compared to illegality, counter-renting allows migrant entrepreneurs partial access to contracts, because migrants can use their local cooperators' business name to sign contracts. Its application is limited, however, because customers may not be willing to enter into a contract where the nominal party is not the actual operator. If a migrant defaults on the contract, it is difficult to go after the local cooperator since he or she is not the actual party to the contract, even though his or her name or seal is on the contract. This form of contract creates extra potential risk and, as a result, is limited in use. While all these legal, illegal, or semi-legal forms involve tremendous costs, relatively speaking, the costs of legal collusion appear to be the lowest.

The costs, however, are not the sole reason why most migrants choose legal collusion. More important, this choice creates a broad consensus or a social norm—that legal collusion is a penumbral behavior acceptable to local authorities (compare with Raymond 2002). Literally speaking, legal collusion is still illegal because, after all, counter-renting is not permitted by the authorities. Yet through this method, migrants at least make an effort to comply with some legal requirements. In this sense, the method lies in a gray area in which only some of the migrants' activities are illegal. From the viewpoint of the migrants, they have paid rent to the locals and have, thus, contributed to the city; therefore, they should not be severely punished by the authorities.³³ From the perspective of the authorities, although migrants do not fully comply with the law, they have at least shown some respect to the authorities. At the end of the day, migrants are just trying to survive in the city so as to achieve an essentially legal objective—developing a business—and legal collusion is not illegal in the same sense that robbery or burglary is illegal. Migrants' efforts to comply with the law in legal collusion, to some extent, are pursued to justify their behavior. Through their efforts, the migrants blur the boundary between what is lawful and what is deviant through their preemptive collaboration with locals. “The boundaries between legal and illegal are stretched out not only by the regulators, but also by the regulated” (McBarnet 1991:323). Their active efforts are indeed “neutralizing” techniques meant to change the impression that officials and locals have of their behavior (Sykes & Matza 1957). Thus, they provide the authorities with an excuse not to check thoroughly if their licenses are authentic or not. Even though officials enjoy a great degree of flexibility in their interpretation of regulations, as has been shown in earlier sections, in reality, legal collusion is often let off more easily than outright illegality.

Most important, legal collusion is facilitated and supported by many institutional players. It is obvious that the local collaborators are the biggest beneficiaries; the rent they receive from migrants is sometimes greater than the income they can earn on their own. Local peasants who rent out extra housing space to migrants increase their income, while marketplace companies that manage counter space see an increased volume in business. Destroying legal collusion completely would also damage the interests of the locals, and they definitely would not be in favor of stricter enforcement. Sporadic campaigns, however, are acceptable for them as they can still receive rent most of the time. Further, sporadic campaigns remind migrants of the importance of being protected by the locals, and they also increase the locals' bargaining power for future collusions (He 2003d).

Even for ordinary citizens in Beijing, legal collusion is not without merit. Developments in the migrant garment business by way of legal collusion solve the problem of garment shortages, which is often seen in planned economies. The prosperous market provides Beijing citizens with more choices and lower garment prices. As consumers of garments, they have no reason to dislike the more competitive market that migrants help create. At the same time, the semi-legal status of migrants reminds ordinary registered citizens of their superior status in the city, in part pacifying their resentment toward migrants and also reinforcing their complacency at possessing the higher social status of urbanites.

Local authorities have no particular reason to strictly enforce the law and destroy legal collusion altogether. They collect some revenue indirectly from migrants: while migrants do not pay taxes or regulation fees directly to the authorities, part of the rent they pay to their collaborating locals will be eventually passed on to the authorities. Law enforcement officials intentionally leave some space for migrant businesses to grow, and they collect bribes and financial gifts from migrants from time to time. Further, migrants bring a prosperous garment market to Beijing, and this increases the popularity of the authorities in general. This is why sporadic campaigns are adopted as a short-term strategy to combat migrants only when pressure from higher-level governments becomes unavoidable. Indeed, legal collusion—almost every migrant engages de facto in one or more illegal transactions—provides authorities with a good excuse to crack down on migrants when the need arises while offering the chance to benefit from them at other times. The degree of legal collusion is eventually controlled by the authorities through campaigns; they can prevent collusion from getting worse, and prevent migrants from overrunning the city. There are, however, other reasons—such as the economic benefits that locals get from the migrants—not to let the circumstances of legal collusion improve. The authorities allow some space for legal collusion to survive in society, but the reason for allowing this is that legal collusion and migrants are useful to locals and to local governments. Hence, there is a kind of equilibrium between legal enforcement and legal collusion (He 2003d).

To put this in another way, the meticulous requirements of license applications not only amount to barriers for migrants to get into the market, but also operate as a panopticon on every aspect of migrants' lives (Foucault 1977). At first glance, this effort has failed because of the resistance and creative strategies of migrants to overcome these barriers, and the authorities do not successfully prevent migration or regulate migrants merely through legal regulations. However, the seeming failure makes the authorities successful regarding social control in general. The authorities balance various interests, maintain social hierarchies, and indirectly create a prosperous market. This is not achieved simply by a set of legal regulations but through a complex system of which the legal regulations are a part. The authorities achieve this despite their failure at preventing migrants from coming into the city. To put it more radically, they succeed through the very failure of the legal barriers preventing migrants from coming to the city.

Conclusions and Implications

This case study addresses why a considerable proportion of rural-urban migrant entrepreneurs do not comply with a piece of licensing legislation in China, a country still in drastic transition from an authoritarian regime to a market-oriented one. In this study, I find that legitimacy, which in most liberal democratic societies is always assumed to be associated with the law, has little influence. Subsequently, instrumental considerations add weight to migrants' decisions on whether or not to comply with the law. Some migrants choose to comply, not because they have any particular respect for the law; while some choose illegality, not because they are bad-natured. They make their choices simply to ensure that their businesses benefit more. Taking half-measures such as legal collusion nonetheless has become a widespread strategy for most migrant entrepreneurs. This strategy is favorable not only because of the instrumental advantages it has against legality or illegality, or because it is more acceptable for the authorities but, most important, because of institutional support for such collaboration. Legal collusion is reproduced in such an institutional environment, despite the increasing important role of law in the society generally. In this case, there is a general equilibrium through which social control is achieved and sustained while simultaneously various interests are satisfied. In this equilibrium, local authorities may not really want to enforce the law, or they may be unable to enforce the law.

This case thus presents a complicated picture of law enforcement in a society where legitimacy of the law is absent. The society is not run simply through coercion and reward, or through instrumental considerations, as often assumed, but through a set of meticulous regulations and sporadic enforcement of the law. The method of social control through this kind of law enforcement is sophisticated, calculated, and successful. Due to the existence of a huge developmental difference between rural and urban China, as a result of the country's long-term policy subsidizing urban areas with the output of the rural areas during the planned-economy period, migrants are still willing to conduct businesses semi-legally in the city. Despite the legal barriers, they believe that they are better off in the city than in their rural hometown. Their low level of social economic status determines their eagerness to seize the market opportunities in the city. At the same time, partly due to their low income level, they can afford the price of noncompliance—that is, the sanction of the state, especially when compared to what they could gain through a semi-legal business. In this sense, they have no better choice than to become the prey of local authorities. The choice between complying and not complying is often conditioned by the institutional structure that prevailed in the pre-transitional period. Thus, the prevalence of semi-illegality is deeply rooted in the historical and socioeconomic setting, and institutional change is likely to be path-dependent (North 1997).

The complicated story recounted here poses challenges for improving the situation that has been described. Widespread legal collusion, which is closely associated with corruption and the practice of rent-seeking, will surely have negative effects on the rule of law and on economic performance, especially in the long run (Rose-Ackerman 1999:21–3). However, making licensing for migrants cheaper, easier, and less discriminatory will definitely encounter resistance because the city may not be able to accommodate more migrants. Local registered citizens and especially local business operators are the greatest beneficiaries of migrant rent and will surely oppose these proposals (Hellman 1998). The local legislature, which more or less represents the interests of its constituents, may not want to improve the situation through legislative surgery. It is very unlikely that the restrictive regulations will be abolished without outside pressure or significant changes in surrounding circumstances.

What are the general implications of this study on legal enforcement? Theoretically, the idea of legal collusion broadens our horizon regarding the field of legal enforcement. Traditional wisdom has realized the difficulty of legal enforcement. Bardach and Kagan, for example, have suggested that law is not always consistently enforced, because there are good or not-so-good administrative inspectors and regulators (1982). On the other hand, the regulated subjects are often able to exploit vague rules and cumbersome processes to evade regulations (Macaulay 1993). Although these studies have greatly deepened our understanding of the difficulties and complexities of legal enforcement, a potential weakness is that they all approach the question by considering the dichotomy between regulatory agencies and regulated subjects. This dichotomy, as useful as it has been, nonetheless limits our focus mostly on the interaction between the regulators and the regulated, which could ignore the interactions among different subgroups of the regulated, even though such interactions could be subversive to legal enforcement. The concept of legal collusion could be a starting point to enlarge researchers' focus on those originally ignored field (He 2004a).

From this perspective, we would find that the legal collusion discussed here is by no means unique to migrants and locals in Beijing; indeed, it exists in other sectors of China and across the world at large. For example, in a fieldwork study of the development of private enterprises in Xiamen, Fujian province, Wank finds that due to the differential treatment toward private enterprises and state-owned enterprises, the expansion of private enterprises is embedded in cooperation and collusion between entrepreneurs and officials of state-owned enterprises (Wank 1993:6). In Indonesia, decades ago,

[A]ttempts to control the Chinese have almost everywhere run into the bewildering maze of overlapping Chinese organizations which exist in every country of the area, and they have been frustrated by Chinese evasion, ability and indispensability…. Licensing systems have been thwarted by the willingness of inexperienced indigenous businessmen, whom they were intended to benefit, to sell their import and export permits to the Chinese who possessed what they lacked—organization, contacts, experience and capital. (Thompson & Adloff 1955:6–7)

Similar legal collusion occurs in the borderland between the United States and Mexico. In his study of illegal Mexican aliens, Stoddard points out that illegal Mexican aliens are actually a “normal, functioning ingredient of Southwestern agribusiness, encouraged and utilized by it with the approval and support of social and cultural institutions of the region with the tacit cooperation of border control agencies and legal authorities” (1976:175). He demonstrates that collusion among employees, political and legal officials, church functionaries, governmental enforcement agencies, and the governments of the two nations all help sustain the use of illegal Mexican labor in border agriculture. Inside America, some affirmative programs have set aside contracts for minority-owned companies, but non-minorities have adopted strategies such as arranging for someone from a minority to act as a “front person.”

Many factors may have contributed to these situations, which are similar to legal collusion. A lack of legitimacy in law itself, a discriminatory treatment toward different kinds of groups, a strong administrative system with wide discretion of law enforcement officials, and a low income level where the regulated and the officials are keen to earn more money are all relevant. A tentative hypothesis is that if most of these conditions are satisfied, legal collusion or the like can be expected.

At the core of these factors, discriminatory legal treatment is a significant and immediate cause for legal collusion. Unfortunately, it is hard to get rid of discriminatory treatment. In our case, we have shown that it may not be feasible to eradicate discriminatory license requirements. The strategies of other scenarios for reducing or tackling legal collusion may vary greatly from one case to another, depending on the specific political and social environment in which it is embedded. One thing is clear, however: it is not going to be easy. Legal collusion is deeply rooted in historical and socioeconomic conditions, and in the political structure as well. Its stubborn persistence demonstrates the difficulty of legal enforcement, and reminds us of the pitfalls of simplistic legal instrumentalism.